Merge branch 'main' into dependabot/github_actions/peter-evans/find-comment-4

This commit is contained in:
aivong-openhands
2026-03-17 17:30:43 -05:00
committed by GitHub
1587 changed files with 148523 additions and 107951 deletions

View File

@@ -0,0 +1,37 @@
---
name: upcoming-release
description: This skill should be used when the user asks to "generate release notes", "list upcoming release PRs", "summarize upcoming release", "/upcoming-release", or needs to know what changes are part of an upcoming release.
---
# Upcoming Release Summary
Generate a concise summary of PRs included in the upcoming release.
## Prerequisites
Two commit SHAs are required:
- **First SHA**: The older commit (current release)
- **Second SHA**: The newer commit (what's being released)
If the user does not provide both SHAs, ask for them before proceeding.
## Workflow
1. Run the script from the repository root with the `--json` flag:
```bash
.github/scripts/find_prs_between_commits.py <older-sha> <newer-sha> --json
```
2. Filter out PRs that are:
- Chores
- Dependency updates
- Adding logs
- Refactors
3. Categorize the remaining PRs:
- **Features** - New functionality
- **Bug fixes** - Corrections to existing behavior
- **Security/CVE fixes** - Security-related changes
- **Other** - Everything else
4. Format the output with PRs listed under their category, including the PR number and a brief description.

View File

@@ -0,0 +1,123 @@
---
name: update-sdk
description: This skill should be used when the user asks to "update SDK", "bump SDK version", "pin SDK to a commit", "test unreleased SDK", "update agent-server image", "bump the version", "prepare a release", "what files change for a release", or needs to know how SDK packages are managed in the OpenHands repository. For detailed reference material, see references/docker-image-locations.md and references/sdk-pinning-examples.md in this skill directory.
---
# Update SDK
Bump SDK packages (`openhands-sdk`, `openhands-agent-server`, `openhands-tools`), pin them to unreleased commits for testing, and cut an OpenHands release.
## Quick Summary — How Many Files Change?
| Activity | Manual edits | Auto-regenerated | Total |
|----------|:------------:|:----------------:|:-----:|
| **SDK bump** (released PyPI version) | 2 | 3 | **5** |
| **SDK pin** (unreleased git commit) | 3 | 3 | **6** |
| **Release commit** (version bump) | 3 | 0 | **3** |
The 3 auto-regenerated files are always: `poetry.lock`, `uv.lock`, `enterprise/poetry.lock`.
## SDK Package Bump — 2 Files + 3 Lock Files
Land as a separate PR before the release. Examples: `929dcc3` (SDK 1.11.5), `cd235cc` (SDK 1.11.4).
| File | What to change |
|------|----------------|
| `pyproject.toml` | `openhands-sdk`, `openhands-agent-server`, `openhands-tools` in **two** sections: the `dependencies` array (PEP 508) **and** `[tool.poetry.dependencies]` |
| `openhands/app_server/sandbox/sandbox_spec_service.py` | `AGENT_SERVER_IMAGE` constant — set to `ghcr.io/openhands/agent-server:<version>-python` |
Then regenerate lock files:
```bash
poetry lock && uv lock && cd enterprise && poetry lock && cd ..
```
## Docker Image Locations — All Hardcoded References
For the complete inventory of every file containing a hardcoded Docker image tag or repository, see `references/docker-image-locations.md`. Key files that must stay in sync during an SDK bump:
| File | Image reference | Updated during SDK bump? |
|------|----------------|:------------------------:|
| `openhands/app_server/sandbox/sandbox_spec_service.py` | `AGENT_SERVER_IMAGE = 'ghcr.io/openhands/agent-server:<tag>-python'` | ✅ Yes |
| `docker-compose.yml` | `AGENT_SERVER_IMAGE_TAG` default | ✅ Should be |
| `containers/dev/compose.yml` | `AGENT_SERVER_IMAGE_REPOSITORY` + `_TAG` defaults | ✅ Should be |
> **CI enforcement:** `.github/workflows/check-version-consistency.yml` validates version consistency and compose file image references on every PR and push to main.
### ⚠️ Docker Image Tag Gotcha (merge-commit SHA)
The SDK CI in `software-agent-sdk` repo tags Docker images with the **GitHub Actions merge-commit SHA**, NOT the PR head-commit SHA. When pinning to an SDK PR branch:
1. Check the SDK PR description for the actual image tag (look for the `AGENT_SERVER_IMAGES` section)
2. Or query the CI logs: the "Consolidate Build Information" job prints `"short_sha": "<tag>"`
3. The merge-commit SHA differs from the head SHA shown in the PR
For released SDK versions, images use a version tag (e.g., `1.12.0-python`) — no merge-commit ambiguity.
## Cutting a Release — 3 Files
A release commit updates the version string across 3 files. Gold-standard examples: 1.3.0 (`d063c8c`), 1.4.0 (`495f48b`).
| File | What to change |
|------|----------------|
| `pyproject.toml` | `version = "X.Y.Z"` under `[tool.poetry]` |
| `frontend/package.json` | `"version": "X.Y.Z"` |
| `frontend/package-lock.json` | `"version": "X.Y.Z"` in **two** places (root object and `packages[""]`) |
> **Note:** `openhands/version.py` reads the version from `pyproject.toml` at runtime — no manual edit needed there.
### Compose Files (2 files)
Both compose files should use `ghcr.io/openhands/agent-server` with the current SDK version tag.
| File | What to verify |
|------|----------------|
| `docker-compose.yml` | `AGENT_SERVER_IMAGE_REPOSITORY` defaults to agent-server, `AGENT_SERVER_IMAGE_TAG` is current |
| `containers/dev/compose.yml` | Same — must use agent-server, not runtime |
### Release Workflow
#### Step 1: Verify the SDK bump has landed
```bash
grep -n "openhands-sdk\|openhands-agent-server\|openhands-tools" pyproject.toml
grep -n "AGENT_SERVER_IMAGE" openhands/app_server/sandbox/sandbox_spec_service.py
grep "AGENT_SERVER_IMAGE_TAG" docker-compose.yml containers/dev/compose.yml
```
#### Step 2: Bump version numbers
```bash
# Edit pyproject.toml, frontend/package.json, frontend/package-lock.json
git add pyproject.toml frontend/package.json frontend/package-lock.json
git commit -m "Release X.Y.Z"
git tag X.Y.Z
```
Create a `saas-rel-X.Y.Z` branch from the tagged commit for the SaaS deployment pipeline.
#### Step 3: CI builds Docker images automatically
The `ghcr-build.yml` workflow triggers on tag pushes and produces:
- `ghcr.io/openhands/openhands:X.Y.Z`, `X.Y`, `X`, `latest`
- `ghcr.io/openhands/runtime:X.Y.Z-nikolaik`, `X.Y-nikolaik`
The tagging logic lives in `containers/build.sh` — when `GITHUB_REF_NAME` matches a semver pattern (`^[0-9]+\.[0-9]+\.[0-9]+$`), it auto-generates major, major.minor, and `latest` tags.
## Development: Pin SDK to an Unreleased Commit
For detailed examples of all pinning formats (commit, branch, uv-only), see `references/sdk-pinning-examples.md`.
### Files to change (3 manual + 3 lock files)
| File | What to change |
|------|----------------|
| `pyproject.toml` | Pin all 3 SDK packages in **both** `dependencies` and `[tool.poetry.dependencies]` |
| `openhands/app_server/sandbox/sandbox_spec_service.py` | `AGENT_SERVER_IMAGE` — use the merge-commit SHA tag, NOT the head-commit SHA |
| `docker-compose.yml` | `AGENT_SERVER_IMAGE_TAG` default (for local development) |
| `poetry.lock` | Auto-regenerated via `poetry lock` |
| `uv.lock` | Auto-regenerated via `uv lock` |
| `enterprise/poetry.lock` | Auto-regenerated via `cd enterprise && poetry lock` |
### CI guard
The `check-package-versions.yml` workflow blocks merging to `main` if `[tool.poetry.dependencies]` contains any `rev` fields. This ensures unreleased SDK pins do not accidentally ship in a release.

View File

@@ -0,0 +1,84 @@
# Docker Image Locations — Complete Inventory
Every file in the OpenHands repository containing a hardcoded Docker image tag, repository, or version-pinned image reference. Organized by update cadence.
## Updated During SDK Bump (must change)
These files contain image tags that **must** be updated whenever the SDK version or pinned commit changes.
### `openhands/app_server/sandbox/sandbox_spec_service.py`
- **Line:** `AGENT_SERVER_IMAGE = 'ghcr.io/openhands/agent-server:<tag>-python'`
- **Format:** `<sdk-version>-python` for releases (e.g., `1.12.0-python`), `<7-char-commit-hash>-python` for dev pins
- **Source of truth** for which agent-server image the app server pulls at runtime
- **⚠️ Gotcha:** When pinning to an SDK PR, the image tag is the **merge-commit SHA** from GitHub Actions, not the PR head-commit SHA. Check the SDK PR description or CI logs for the correct tag.
### `docker-compose.yml`
- **Lines:**
```yaml
- AGENT_SERVER_IMAGE_REPOSITORY=${AGENT_SERVER_IMAGE_REPOSITORY:-ghcr.io/openhands/agent-server}
- AGENT_SERVER_IMAGE_TAG=${AGENT_SERVER_IMAGE_TAG:-<tag>-python}
```
- Used by `docker compose up` for local development
### `containers/dev/compose.yml`
- **Lines:**
```yaml
- AGENT_SERVER_IMAGE_REPOSITORY=${AGENT_SERVER_IMAGE_REPOSITORY:-ghcr.io/openhands/agent-server}
- AGENT_SERVER_IMAGE_TAG=${AGENT_SERVER_IMAGE_TAG:-<tag>-python}
```
- Used by the dev container setup
- **Known issue:** On main as of 1.4.0, this file still points to `ghcr.io/openhands/runtime` instead of `agent-server`, and the tag is `1.2-nikolaik` (stale from the V0 era). The `check-version-consistency.yml` CI workflow catches this.
## Updated During Release Commit (version string only)
### `pyproject.toml`
- **Line:** `version = "X.Y.Z"` under `[tool.poetry]`
- The Python version is derived from this at runtime via `openhands/version.py`
### `frontend/package.json`
- **Line:** `"version": "X.Y.Z"`
### `frontend/package-lock.json`
- **Two places:** root `"version": "X.Y.Z"` and `packages[""].version`
## Dynamic References (auto-derived, no manual update)
### `openhands/version.py`
- Reads version from `pyproject.toml` at runtime → `openhands.__version__`
### `openhands/resolver/issue_resolver.py`
- Builds `ghcr.io/openhands/runtime:{openhands.__version__}-nikolaik` dynamically
### `openhands/runtime/utils/runtime_build.py`
- Base repo URL `ghcr.io/openhands/runtime` is a constant; version comes from elsewhere
### `.github/scripts/update_pr_description.sh`
- Uses `${SHORT_SHA}` variable at CI runtime, not hardcoded
### `enterprise/Dockerfile`
- `ARG BASE="ghcr.io/openhands/openhands"` — base image, version supplied at build time
## V0 Legacy Files (separate update cadence)
These reference the V0 runtime image (`ghcr.io/openhands/runtime:X.Y-nikolaik`) for local Docker/Kubernetes paths. They are **not** updated as part of a V1 release but may be updated independently.
### `Development.md`
- `export SANDBOX_RUNTIME_CONTAINER_IMAGE=ghcr.io/openhands/runtime:X.Y-nikolaik`
### `openhands/runtime/impl/kubernetes/README.md`
- `runtime_container_image = "docker.openhands.dev/openhands/runtime:X.Y-nikolaik"`
### `enterprise/enterprise_local/README.md`
- Uses `ghcr.io/openhands/runtime:main-nikolaik` (points to `main`, not versioned)
### `third_party/runtime/impl/daytona/README.md`
- Uses `${OPENHANDS_VERSION}` variable, not hardcoded
## Image Registries
| Registry | Usage |
|----------|-------|
| `ghcr.io/openhands/agent-server` | V1 agent-server (sandbox) — built by SDK repo CI |
| `ghcr.io/openhands/openhands` | Main app image — built by `ghcr-build.yml` |
| `ghcr.io/openhands/runtime` | V0 runtime sandbox — built by `ghcr-build.yml` |
| `docker.openhands.dev/openhands/*` | Mirror/CDN for the above images |

View File

@@ -0,0 +1,103 @@
# SDK Pinning Examples
Examples from real commits showing how to pin SDK packages to unreleased commits, branches, or released versions.
## Pin to a Specific Commit
Example from commit `169fb76` (pinning all 3 packages to SDK commit `100e9af`):
### `dependencies` array (PEP 508 format)
```toml
"openhands-agent-server @ git+https://github.com/OpenHands/software-agent-sdk.git@100e9af#subdirectory=openhands-agent-server",
"openhands-sdk @ git+https://github.com/OpenHands/software-agent-sdk.git@100e9af#subdirectory=openhands-sdk",
"openhands-tools @ git+https://github.com/OpenHands/software-agent-sdk.git@100e9af#subdirectory=openhands-tools",
```
### `[tool.poetry.dependencies]` (Poetry format)
```toml
openhands-sdk = { git = "https://github.com/OpenHands/software-agent-sdk.git", rev = "100e9af", subdirectory = "openhands-sdk" }
openhands-agent-server = { git = "https://github.com/OpenHands/software-agent-sdk.git", rev = "100e9af", subdirectory = "openhands-agent-server" }
openhands-tools = { git = "https://github.com/OpenHands/software-agent-sdk.git", rev = "100e9af", subdirectory = "openhands-tools" }
```
### `openhands/app_server/sandbox/sandbox_spec_service.py`
```python
AGENT_SERVER_IMAGE = 'ghcr.io/openhands/agent-server:<merge-commit-sha>-python'
```
**⚠️ Important:** The image tag is the **merge-commit SHA** from the SDK CI, not the commit hash used in `pyproject.toml`. Look up the correct tag from the SDK PR description or CI logs.
## Pin to a Branch
Example from commit `430ee1c` (pinning to branch `openhands/issue-2228-sdk-settings-schema`):
### `[tool.poetry.dependencies]`
```toml
openhands-sdk = { git = "https://github.com/OpenHands/software-agent-sdk.git", branch = "openhands/issue-2228-sdk-settings-schema", subdirectory = "openhands-sdk" }
openhands-agent-server = { git = "https://github.com/OpenHands/software-agent-sdk.git", branch = "openhands/issue-2228-sdk-settings-schema", subdirectory = "openhands-agent-server" }
openhands-tools = { git = "https://github.com/OpenHands/software-agent-sdk.git", branch = "openhands/issue-2228-sdk-settings-schema", subdirectory = "openhands-tools" }
```
## Using `[tool.uv.sources]` Override
When only `uv` needs the override (keep PyPI versions in the main arrays), add a `[tool.uv.sources]` section. Example from commit `1daca49`:
```toml
[tool.uv.sources]
openhands-sdk = { git = "https://github.com/OpenHands/software-agent-sdk.git", subdirectory = "openhands-sdk", rev = "4170cca" }
openhands-agent-server = { git = "https://github.com/OpenHands/software-agent-sdk.git", subdirectory = "openhands-agent-server", rev = "4170cca" }
openhands-tools = { git = "https://github.com/OpenHands/software-agent-sdk.git", subdirectory = "openhands-tools", rev = "4170cca" }
```
## Released PyPI Version (standard release)
Example from commit `929dcc3` (SDK 1.11.5):
### `dependencies` array
```toml
"openhands-agent-server==1.11.5",
"openhands-sdk==1.11.5",
"openhands-tools==1.11.5",
```
### `[tool.poetry.dependencies]`
```toml
openhands-sdk = "1.11.5"
openhands-agent-server = "1.11.5"
openhands-tools = "1.11.5"
```
### `openhands/app_server/sandbox/sandbox_spec_service.py`
For released versions, the image tag uses the version number:
```python
AGENT_SERVER_IMAGE = 'ghcr.io/openhands/agent-server:1.11.5-python'
```
However, **some releases use a commit-hash tag** even for the released version. Check which tag format exists on GHCR. Example from `929dcc3`:
```python
AGENT_SERVER_IMAGE = 'ghcr.io/openhands/agent-server:010e847-python'
```
## Regenerate Lock Files
After any change to `pyproject.toml`, always regenerate:
```bash
poetry lock
uv lock
cd enterprise && poetry lock && cd ..
```
## CI Guards
- **`check-package-versions.yml`**: Blocks merge to `main` if `[tool.poetry.dependencies]` contains `rev` fields (prevents shipping unreleased SDK pins)
- **`check-version-consistency.yml`**: Validates version strings match across `pyproject.toml`, `package.json`, `package-lock.json`, and verifies compose files use `agent-server` images

View File

@@ -5,52 +5,113 @@ labels: ['bug']
body:
- type: markdown
attributes:
value: Thank you for taking the time to fill out this bug report. Please provide as much information as possible
to help us understand and address the issue effectively.
value: |
## Thank you for reporting a bug! 🐛
**Please fill out all required fields.** Issues missing critical information (version, installation method, reproduction steps, etc.) will be delayed or closed until complete details are provided.
Clear, detailed reports help us resolve issues faster.
- type: checkboxes
attributes:
label: Is there an existing issue for the same bug? (If one exists, thumbs up or comment on the issue instead).
description: Please check if an issue already exists for the bug you encountered.
label: Is there an existing issue for the same bug?
description: Please search existing issues before creating a new one. If found, react or comment to the duplicate issue instead of making a new one.
options:
- label: I have checked the existing issues.
- label: I have searched existing issues and this is not a duplicate.
required: true
- type: textarea
id: bug-description
attributes:
label: Describe the bug and reproduction steps
description: Provide a description of the issue along with any reproduction steps.
label: Bug Description
description: Clearly describe what went wrong. Be specific and concise.
placeholder: Example - "When I run a Python task, OpenHands crashes after 30 seconds with a connection timeout error."
validations:
required: true
- type: textarea
id: expected-behavior
attributes:
label: Expected Behavior
description: What did you expect to happen?
placeholder: Example - "OpenHands should execute the Python script and return results."
validations:
required: false
- type: textarea
id: actual-behavior
attributes:
label: Actual Behavior
description: What actually happened?
placeholder: Example - "Connection timed out after 30 seconds, task failed with error code 500."
validations:
required: false
- type: textarea
id: reproduction-steps
attributes:
label: Steps to Reproduce
description: Provide clear, step-by-step instructions to reproduce the bug.
placeholder: |
1. Install OpenHands using Docker
2. Configure with Claude 3.5 Sonnet
3. Run command: `openhands run "write a python script"`
4. Wait 30 seconds
5. Error appears
validations:
required: false
- type: dropdown
id: installation
attributes:
label: OpenHands Installation
label: OpenHands Installation Method
description: How are you running OpenHands?
options:
- Docker command in README
- GitHub resolver
- CLI (uv tool install)
- CLI (executable binary)
- CLI (Docker)
- Local GUI (Docker web interface)
- OpenHands Cloud (app.all-hands.dev)
- SDK (Python library)
- Development workflow
- CLI
- app.all-hands.dev
- Other
default: 0
validations:
required: false
- type: input
id: installation-other
attributes:
label: If you selected "Other", please specify
description: Describe your installation method
placeholder: ex. Custom Kubernetes deployment, pip install from source, etc.
- type: input
id: openhands-version
attributes:
label: OpenHands Version
description: What version of OpenHands are you using?
placeholder: ex. 0.9.8, main, etc.
description: What version are you using? Find this in settings or by running `openhands --version`
placeholder: ex. 0.9.8, main, commit hash, etc.
validations:
required: false
- type: checkboxes
id: version-confirmation
attributes:
label: Version Confirmation
description: Bugs on older versions may already be fixed. Please upgrade before submitting.
options:
- label: "I have confirmed this bug exists on the LATEST version of OpenHands"
required: false
- type: input
id: model-name
attributes:
label: Model Name
description: What model are you using?
placeholder: ex. gpt-4o, claude-3-5-sonnet, openrouter/deepseek-r1, etc.
description: Which LLM model are you using?
placeholder: ex. gpt-4o, claude-3-5-sonnet-20241022, openrouter/deepseek-r1, etc.
validations:
required: false
- type: dropdown
id: os
@@ -60,12 +121,46 @@ body:
- MacOS
- Linux
- WSL on Windows
- Windows (Docker Desktop)
- Other
validations:
required: false
- type: input
id: browser
attributes:
label: Browser (if using web UI)
description: |
If applicable, which browser and version?
placeholder: ex. Chrome 131, Firefox 133, Safari 17.2
- type: textarea
id: logs
attributes:
label: Logs and Error Messages
description: |
**Paste relevant logs, error messages, or stack traces.** Use code blocks (```) for formatting.
LLM logs are in `logs/llm/default/`. Include timestamps if errors occurred at a specific time.
placeholder: |
```
Paste error logs here
```
- type: textarea
id: additional-context
attributes:
label: Logs, Errors, Screenshots, and Additional Context
description: Please provide any additional information you think might help. If you want to share the chat history
you can click the thumbs-down (👎) button above the input field and you will get a shareable link
(you can also click thumbs up when things are going well of course!). LLM logs will be stored in the
`logs/llm/default` folder. Please add any additional context about the problem here.
label: Screenshots and Additional Context
description: |
Add screenshots, videos, runtime environment, or other context that helps explain the issue.
💡 **Share conversation history:** In the OpenHands chat UI, click the 👎 or 👍 button (above the message input) to generate a shareable link to your conversation.
placeholder: Drag and drop screenshots here, paste links, or add additional context.
- type: markdown
attributes:
value: |
---
**Note:** Issues with incomplete information may be closed or deprioritized. Maintainers and community members have limited bandwidth and prioritize well-documented bugs that are easier to reproduce and fix. Thank you for your understanding!

2
.github/ISSUE_TEMPLATE/config.yml vendored Normal file
View File

@@ -0,0 +1,2 @@
# disable blank issue creation
blank_issues_enabled: false

View File

@@ -1,17 +0,0 @@
---
name: Feature Request or Enhancement
about: Suggest an idea for an OpenHands feature or enhancement
title: ''
labels: 'enhancement'
assignees: ''
---
**What problem or use case are you trying to solve?**
**Describe the UX or technical implementation you have in mind**
**Additional context**
### If you find this feature request or enhancement useful, make sure to add a 👍 to the issue

View File

@@ -0,0 +1,105 @@
name: Feature Request or Enhancement
description: Suggest a new feature or improvement for OpenHands
title: '[Feature]: '
labels: ['enhancement']
body:
- type: markdown
attributes:
value: |
## Thank you for suggesting a feature! 💡
**Please provide detailed information.** Vague or low-effort requests may be closed. Well-documented feature requests with strong community support are more likely to be added to the roadmap.
- type: checkboxes
attributes:
label: Is there an existing feature request for this?
description: Please search existing issues and feature requests before creating a new one. If found, react or comment to the duplicate issue instead of making a new one.
options:
- label: I have searched existing issues and feature requests, and this is not a duplicate.
required: true
- type: textarea
id: problem-statement
attributes:
label: Problem or Use Case
description: What problem are you trying to solve? What use case would this feature enable?
placeholder: |
Example - "As a developer working on large codebases, I need to search across multiple files simultaneously. Currently, I have to search file-by-file which is time-consuming and inefficient."
validations:
required: true
- type: textarea
id: proposed-solution
attributes:
label: Proposed Solution
description: Describe your ideal solution. What should this feature do? How should it work?
placeholder: |
Example - "Add a global search feature that allows searching across all files in the workspace. Results should show file name, line number, and context around matches. Include regex support and filtering options."
validations:
required: true
- type: textarea
id: alternatives
attributes:
label: Alternatives Considered
description: Have you considered any alternative solutions or workarounds? What are their limitations?
placeholder: Example - "I tried using grep in the terminal, but it's not integrated with the UI and doesn't provide click-to-navigate functionality."
- type: dropdown
id: priority
attributes:
label: Priority / Severity
description: How important is this feature to your workflow?
options:
- "Critical - Blocking my work, no workaround available"
- "High - Significant impact on productivity"
- "Medium - Would improve experience"
- "Low - Nice to have"
default: 2
validations:
required: true
- type: dropdown
id: scope
attributes:
label: Estimated Scope
description: To the best of your knowledge, how complex do you think this feature would be to implement?
options:
- "Small - UI tweak, config option, or minor change"
- "Medium - New feature with moderate complexity"
- "Large - Significant feature requiring architecture changes"
- "Unknown - Not sure about the technical complexity"
default: 3
- type: dropdown
id: feature-area
attributes:
label: Feature Area
description: Which part of OpenHands does this feature relate to? If you select "Other", please specify the area in the Additional Context section below.
options:
- "Agent / AI behavior"
- "User Interface / UX"
- "CLI / Command-line interface"
- "File system / Workspace management"
- "Configuration / Settings"
- "Integrations (GitHub, GitLab, etc.)"
- "Performance / Optimization"
- "Documentation"
- "Other"
validations:
required: true
- type: textarea
id: technical-details
attributes:
label: Technical Implementation Ideas (Optional)
description: If you have technical expertise, share implementation ideas, API suggestions, or relevant technical details.
placeholder: |
Example - "Could use ripgrep library for fast search. Expose results via /api/search endpoint. Frontend can use virtualized list for rendering large result sets."
- type: textarea
id: additional-context
attributes:
label: Additional Context
description: Add any other context, screenshots, mockups, or examples that help illustrate this feature request.
placeholder: Drag and drop screenshots, mockups, or links here.

View File

@@ -1,10 +1,16 @@
<!-- If you are still working on the PR, please mark it as draft. Maintainers will review PRs marked ready for review, which leads to lost time if your PR is actually not ready yet. Keep the PR marked as draft until it is finally ready for review -->
## Summary of PR
<!-- Summarize what the PR does, explaining any non-trivial design decisions. -->
<!-- Summarize what the PR does -->
## Demo Screenshots/Videos
<!-- AI/LLM AGENTS: This section is intended for a human author to add screenshots or videos demonstrating the PR in action (optional). While many pull requests may be generated by AI/LLM agents, we are fine with this as long as a human author has reviewed and tested the changes to ensure accuracy and functionality. -->
## Change Type
<!-- Choose the types that apply to your PR and remove the rest. -->
<!-- Choose the types that apply to your PR -->
- [ ] Bug fix
- [ ] New feature

View File

@@ -0,0 +1,330 @@
#!/usr/bin/env python3
"""
Find all PRs that went in between two commits in the OpenHands/OpenHands repository.
Handles cherry-picks and different merge strategies.
This script is designed to run from within the OpenHands repository under .github/scripts:
.github/scripts/find_prs_between_commits.py
Usage: find_prs_between_commits <older_commit> <newer_commit> [--repo <path>]
"""
import json
import os
import re
import subprocess
import sys
from collections import defaultdict
from pathlib import Path
from typing import Optional
def find_openhands_repo() -> Optional[Path]:
"""
Find the OpenHands repository.
Since this script is designed to live in .github/scripts/, it assumes
the repository root is two levels up from the script location.
Tries:
1. Repository root (../../ from script location)
2. Current directory
3. Environment variable OPENHANDS_REPO
"""
# Check repository root (assuming script is in .github/scripts/)
script_dir = Path(__file__).parent.absolute()
repo_root = (
script_dir.parent.parent
) # Go up two levels: scripts -> .github -> repo root
if (repo_root / '.git').exists():
return repo_root
# Check current directory
if (Path.cwd() / '.git').exists():
return Path.cwd()
# Check environment variable
if 'OPENHANDS_REPO' in os.environ:
repo_path = Path(os.environ['OPENHANDS_REPO'])
if (repo_path / '.git').exists():
return repo_path
return None
def run_git_command(cmd: list[str], repo_path: Path) -> str:
"""Run a git command in the repository directory and return its output."""
try:
result = subprocess.run(
cmd, capture_output=True, text=True, check=True, cwd=str(repo_path)
)
return result.stdout.strip()
except subprocess.CalledProcessError as e:
print(f'Error running git command: {" ".join(cmd)}', file=sys.stderr)
print(f'Error: {e.stderr}', file=sys.stderr)
sys.exit(1)
def extract_pr_numbers_from_message(message: str) -> set[int]:
"""Extract PR numbers from commit message in any common format."""
# Match #12345 anywhere, including in patterns like (#12345) or "Merge pull request #12345"
matches = re.findall(r'#(\d+)', message)
return set(int(m) for m in matches)
def get_commit_info(commit_hash: str, repo_path: Path) -> tuple[str, str, str]:
"""Get commit subject, body, and author from a commit hash."""
subject = run_git_command(
['git', 'log', '-1', '--format=%s', commit_hash], repo_path
)
body = run_git_command(['git', 'log', '-1', '--format=%b', commit_hash], repo_path)
author = run_git_command(
['git', 'log', '-1', '--format=%an <%ae>', commit_hash], repo_path
)
return subject, body, author
def get_commits_between(
older_commit: str, newer_commit: str, repo_path: Path
) -> list[str]:
"""Get all commit hashes between two commits."""
commits_output = run_git_command(
['git', 'rev-list', f'{older_commit}..{newer_commit}'], repo_path
)
if not commits_output:
return []
return commits_output.split('\n')
def get_pr_info_from_github(pr_number: int, repo_path: Path) -> Optional[dict]:
"""Get PR information from GitHub API if GITHUB_TOKEN is available."""
try:
# Set up environment with GitHub token
env = os.environ.copy()
if 'GITHUB_TOKEN' in env:
env['GH_TOKEN'] = env['GITHUB_TOKEN']
result = subprocess.run(
[
'gh',
'pr',
'view',
str(pr_number),
'--json',
'number,title,author,mergedAt,baseRefName,headRefName,url',
],
capture_output=True,
text=True,
check=True,
env=env,
cwd=str(repo_path),
)
return json.loads(result.stdout)
except (subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
return None
def find_prs_between_commits(
older_commit: str, newer_commit: str, repo_path: Path
) -> dict[int, dict]:
"""
Find all PRs that went in between two commits.
Returns a dictionary mapping PR numbers to their information.
"""
print(f'Repository: {repo_path}', file=sys.stderr)
print('Finding PRs between commits:', file=sys.stderr)
print(f' Older: {older_commit}', file=sys.stderr)
print(f' Newer: {newer_commit}', file=sys.stderr)
print(file=sys.stderr)
# Verify commits exist
try:
run_git_command(['git', 'rev-parse', '--verify', older_commit], repo_path)
run_git_command(['git', 'rev-parse', '--verify', newer_commit], repo_path)
except SystemExit:
print('Error: One or both commits not found in repository', file=sys.stderr)
sys.exit(1)
# Extract PRs from the older commit itself (to exclude from results)
# These PRs are already included at or before the older commit
older_subject, older_body, _ = get_commit_info(older_commit, repo_path)
older_message = f'{older_subject}\n{older_body}'
excluded_prs = extract_pr_numbers_from_message(older_message)
if excluded_prs:
print(
f'Excluding PRs already in older commit: {", ".join(f"#{pr}" for pr in sorted(excluded_prs))}',
file=sys.stderr,
)
print(file=sys.stderr)
# Get all commits between the two
commits = get_commits_between(older_commit, newer_commit, repo_path)
print(f'Found {len(commits)} commits to analyze', file=sys.stderr)
print(file=sys.stderr)
# Extract PR numbers from all commits
pr_info: dict[int, dict] = {}
commits_by_pr: dict[int, list[str]] = defaultdict(list)
for commit_hash in commits:
subject, body, author = get_commit_info(commit_hash, repo_path)
full_message = f'{subject}\n{body}'
pr_numbers = extract_pr_numbers_from_message(full_message)
for pr_num in pr_numbers:
# Skip PRs that are already in the older commit
if pr_num in excluded_prs:
continue
commits_by_pr[pr_num].append(commit_hash)
if pr_num not in pr_info:
pr_info[pr_num] = {
'number': pr_num,
'first_commit': commit_hash[:8],
'first_commit_subject': subject,
'commits': [],
'github_info': None,
}
pr_info[pr_num]['commits'].append(
{'hash': commit_hash[:8], 'subject': subject, 'author': author}
)
# Try to get additional info from GitHub API
print('Fetching additional info from GitHub API...', file=sys.stderr)
for pr_num in pr_info.keys():
github_info = get_pr_info_from_github(pr_num, repo_path)
if github_info:
pr_info[pr_num]['github_info'] = github_info
print(file=sys.stderr)
return pr_info
def print_results(pr_info: dict[int, dict]):
"""Print the results in a readable format."""
sorted_prs = sorted(pr_info.items(), key=lambda x: x[0])
print(f'{"=" * 80}')
print(f'Found {len(sorted_prs)} PRs')
print(f'{"=" * 80}')
print()
for pr_num, info in sorted_prs:
print(f'PR #{pr_num}')
if info['github_info']:
gh = info['github_info']
print(f' Title: {gh["title"]}')
print(f' Author: {gh["author"]["login"]}')
print(f' URL: {gh["url"]}')
if gh.get('mergedAt'):
print(f' Merged: {gh["mergedAt"]}')
if gh.get('baseRefName'):
print(f' Base: {gh["baseRefName"]}{gh["headRefName"]}')
else:
print(f' Subject: {info["first_commit_subject"]}')
# Show if this PR has multiple commits (cherry-picked or multiple commits)
commit_count = len(info['commits'])
if commit_count > 1:
print(
f' ⚠️ Found {commit_count} commits (possible cherry-pick or multi-commit PR):'
)
for commit in info['commits'][:3]: # Show first 3
print(f' {commit["hash"]}: {commit["subject"][:60]}')
if commit_count > 3:
print(f' ... and {commit_count - 3} more')
else:
print(f' Commit: {info["first_commit"]}')
print()
def main():
if len(sys.argv) < 3:
print('Usage: find_prs_between_commits <older_commit> <newer_commit> [options]')
print()
print('Arguments:')
print(' <older_commit> The older commit hash (or ref)')
print(' <newer_commit> The newer commit hash (or ref)')
print()
print('Options:')
print(' --json Output results in JSON format')
print(' --repo <path> Path to OpenHands repository (default: auto-detect)')
print()
print('Example:')
print(
' find_prs_between_commits c79e0cd3c7a2501a719c9296828d7a31e4030585 35bddb14f15124a3dc448a74651a6592911d99e9'
)
print()
print('Repository Detection:')
print(' The script will try to find the OpenHands repository in this order:')
print(' 1. --repo argument')
print(' 2. Repository root (../../ from script location)')
print(' 3. Current directory')
print(' 4. OPENHANDS_REPO environment variable')
print()
print('Environment variables:')
print(
' GITHUB_TOKEN Optional. If set, will fetch additional PR info from GitHub API'
)
print(' OPENHANDS_REPO Optional. Path to OpenHands repository')
sys.exit(1)
older_commit = sys.argv[1]
newer_commit = sys.argv[2]
json_output = '--json' in sys.argv
# Check for --repo argument
repo_path = None
if '--repo' in sys.argv:
repo_idx = sys.argv.index('--repo')
if repo_idx + 1 < len(sys.argv):
repo_path = Path(sys.argv[repo_idx + 1])
if not (repo_path / '.git').exists():
print(f'Error: {repo_path} is not a git repository', file=sys.stderr)
sys.exit(1)
# Auto-detect repository if not specified
if repo_path is None:
repo_path = find_openhands_repo()
if repo_path is None:
print('Error: Could not find OpenHands repository', file=sys.stderr)
print('Please either:', file=sys.stderr)
print(
' 1. Place this script in .github/scripts/ within the OpenHands repository',
file=sys.stderr,
)
print(' 2. Run from the OpenHands repository directory', file=sys.stderr)
print(
' 3. Use --repo <path> to specify the repository location',
file=sys.stderr,
)
print(' 4. Set OPENHANDS_REPO environment variable', file=sys.stderr)
sys.exit(1)
# Find PRs
pr_info = find_prs_between_commits(older_commit, newer_commit, repo_path)
if json_output:
# Output as JSON
print(json.dumps(pr_info, indent=2))
else:
# Print results in human-readable format
print_results(pr_info)
# Also print a simple list for easy copying
print(f'{"=" * 80}')
print('PR Numbers (for easy copying):')
print(f'{"=" * 80}')
sorted_pr_nums = sorted(pr_info.keys())
print(', '.join(f'#{pr}' for pr in sorted_pr_nums))
if __name__ == '__main__':
main()

View File

@@ -0,0 +1,122 @@
name: Check Version Consistency
on:
push:
branches: [main]
pull_request:
workflow_dispatch:
jobs:
check-version-consistency:
runs-on: ubuntu-latest
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Set up Python
uses: actions/setup-python@v6
with:
python-version: "3.12"
- name: Check version and Docker image tag consistency
run: |
python - <<'PY'
import json
import re
import sys
import tomllib
errors = []
warnings = []
# ── 1. Extract the canonical version from pyproject.toml ──────────
with open("pyproject.toml", "rb") as f:
pyproject = tomllib.load(f)
version = pyproject["tool"]["poetry"]["version"]
major_minor = ".".join(version.split(".")[:2])
print(f"📦 pyproject.toml version: {version} (major.minor: {major_minor})")
# ── 2. Check frontend/package.json ────────────────────────────────
with open("frontend/package.json") as f:
pkg = json.load(f)
if pkg["version"] != version:
errors.append(
f"frontend/package.json version is '{pkg['version']}', expected '{version}'"
)
else:
print(f" ✔ frontend/package.json: {pkg['version']}")
# ── 3. Check frontend/package-lock.json (2 places) ───────────────
with open("frontend/package-lock.json") as f:
lock = json.load(f)
for key, val in [
("root.version", lock.get("version")),
('packages[""].version', lock.get("packages", {}).get("", {}).get("version")),
]:
if val != version:
errors.append(
f"frontend/package-lock.json {key} is '{val}', expected '{version}'"
)
else:
print(f" ✔ frontend/package-lock.json {key}: {val}")
# ── 4. Check compose files use agent-server images ─────────────────
# Both compose files should use ghcr.io/.../agent-server (not runtime).
# Agent-server tags use SDK version (e.g. "1.12.0-python") or commit
# hashes (e.g. "31536c8-python") — both are acceptable.
repo_pattern = re.compile(r"AGENT_SERVER_IMAGE_REPOSITORY[^}]*:-([^}]+)")
tag_pattern = re.compile(r"AGENT_SERVER_IMAGE_TAG:-([^}]+)")
for filepath in ["docker-compose.yml", "containers/dev/compose.yml"]:
try:
with open(filepath) as f:
content = f.read()
except FileNotFoundError:
warnings.append(f"{filepath}: file not found")
continue
repos = repo_pattern.findall(content)
tags = tag_pattern.findall(content)
if not repos:
warnings.append(f"{filepath}: no AGENT_SERVER_IMAGE_REPOSITORY default found")
else:
repo = repos[0]
if "agent-server" not in repo:
errors.append(
f"{filepath}: AGENT_SERVER_IMAGE_REPOSITORY defaults to '{repo}', "
f"expected an agent-server image (not runtime)"
)
else:
print(f" ✔ {filepath} image repository: {repo}")
if not tags:
warnings.append(f"{filepath}: no AGENT_SERVER_IMAGE_TAG default found")
else:
tag = tags[0]
if not tag:
errors.append(f"{filepath}: AGENT_SERVER_IMAGE_TAG default is empty")
else:
print(f" ✔ {filepath} image tag: {tag}")
# ── 5. Report ─────────────────────────────────────────────────────
print()
if warnings:
print("⚠ Warnings:")
for w in warnings:
print(f" {w}")
print()
if errors:
print("❌ FAILED: Version inconsistencies found:\n")
for e in errors:
print(f" ✖ {e}")
print(
"\nAll version numbers and Docker image tags must be consistent."
"\nSee .agents/skills/update-sdk/SKILL.md for the full checklist."
)
sys.exit(1)
else:
print("✅ All version numbers and Docker image tags are consistent.")
PY

View File

@@ -43,7 +43,7 @@ jobs:
⚠️ This PR contains **migrations**
- name: Comment warning on PR
uses: peter-evans/create-or-update-comment@v4
uses: peter-evans/create-or-update-comment@v5
with:
issue-number: ${{ github.event.pull_request.number }}
comment-id: ${{ steps.find-comment.outputs.comment-id }}

View File

@@ -1,29 +0,0 @@
# Feature branch preview for enterprise code
name: Enterprise Preview
# Run on PRs labeled
on:
pull_request:
types: [labeled]
# Match ghcr-build.yml, but don't interrupt it.
concurrency:
group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
cancel-in-progress: false
jobs:
# This must happen for the PR Docker workflow when the label is present,
# and also if it's added after the fact. Thus, it exists in both places.
enterprise-preview:
name: Enterprise preview
if: github.event.label.name == 'deploy'
runs-on: blacksmith-4vcpu-ubuntu-2204
steps:
# This should match the version in ghcr-build.yml
- name: Trigger remote job
run: |
curl --fail-with-body -sS -X POST \
-H "Authorization: Bearer ${{ secrets.PAT_TOKEN }}" \
-H "Accept: application/vnd.github+json" \
-d "{\"ref\": \"main\", \"inputs\": {\"openhandsPrNumber\": \"${{ github.event.pull_request.number }}\", \"deployEnvironment\": \"feature\", \"enterpriseImageTag\": \"pr-${{ github.event.pull_request.number }}\" }}" \
https://api.github.com/repos/OpenHands/deploy/actions/workflows/deploy.yaml/dispatches

View File

@@ -9,6 +9,7 @@ on:
push:
branches:
- main
- "saas-rel-*"
tags:
- "*"
pull_request:
@@ -39,8 +40,7 @@ jobs:
run: |
if [[ "$GITHUB_EVENT_NAME" == "pull_request" ]]; then
json=$(jq -n -c '[
{ image: "nikolaik/python-nodejs:python3.12-nodejs22", tag: "nikolaik" },
{ image: "ubuntu:24.04", tag: "ubuntu" }
{ image: "nikolaik/python-nodejs:python3.12-nodejs22", tag: "nikolaik" }
]')
else
json=$(jq -n -c '[
@@ -64,7 +64,7 @@ jobs:
with:
ref: ${{ github.event.pull_request.head.sha }}
- name: Set up QEMU
uses: docker/setup-qemu-action@v3.6.0
uses: docker/setup-qemu-action@v3.7.0
with:
image: tonistiigi/binfmt:latest
- name: Login to GHCR
@@ -102,7 +102,7 @@ jobs:
with:
ref: ${{ github.event.pull_request.head.sha }}
- name: Set up QEMU
uses: docker/setup-qemu-action@v3.6.0
uses: docker/setup-qemu-action@v3.7.0
with:
image: tonistiigi/binfmt:latest
- name: Login to GHCR
@@ -149,6 +149,9 @@ jobs:
push: true
tags: ${{ env.DOCKER_TAGS }}
platforms: ${{ env.DOCKER_PLATFORM }}
# Caching directives to boost performance
cache-from: type=registry,ref=ghcr.io/${{ env.REPO_OWNER }}/runtime:buildcache-${{ matrix.base_image.tag }}
cache-to: type=registry,ref=ghcr.io/${{ env.REPO_OWNER }}/runtime:buildcache-${{ matrix.base_image.tag }},mode=max
build-args: ${{ env.DOCKER_BUILD_ARGS }}
context: containers/runtime
provenance: false
@@ -237,166 +240,15 @@ jobs:
# Add build attestations for better security
sbom: true
enterprise-preview:
name: Enterprise preview
if: github.event_name == 'pull_request' && contains(github.event.pull_request.labels.*.name, 'deploy')
runs-on: blacksmith-4vcpu-ubuntu-2204
needs: [ghcr_build_enterprise]
steps:
# This should match the version in enterprise-preview.yml
- name: Trigger remote job
run: |
curl --fail-with-body -sS -X POST \
-H "Authorization: Bearer ${{ secrets.PAT_TOKEN }}" \
-H "Accept: application/vnd.github+json" \
-d "{\"ref\": \"main\", \"inputs\": {\"openhandsPrNumber\": \"${{ github.event.pull_request.number }}\", \"deployEnvironment\": \"feature\", \"enterpriseImageTag\": \"pr-${{ github.event.pull_request.number }}\" }}" \
https://api.github.com/repos/OpenHands/deploy/actions/workflows/deploy.yaml/dispatches
# Run unit tests with the Docker runtime Docker images as root
test_runtime_root:
name: RT Unit Tests (Root)
needs: [ghcr_build_runtime, define-matrix]
runs-on: blacksmith-4vcpu-ubuntu-2404
strategy:
fail-fast: false
matrix:
base_image: ${{ fromJson(needs.define-matrix.outputs.base_image) }}
steps:
- uses: actions/checkout@v4
- name: Set up Docker Buildx
id: buildx
uses: docker/setup-buildx-action@v3
- name: Download runtime source for fork
if: github.event.pull_request.head.repo.fork
uses: actions/download-artifact@v6
with:
name: runtime-src-${{ matrix.base_image.tag }}
path: containers/runtime
- name: Lowercase Repository Owner
run: |
echo REPO_OWNER=$(echo ${{ github.repository_owner }} | tr '[:upper:]' '[:lower:]') >> $GITHUB_ENV
# Forked repos can't push to GHCR, so we need to rebuild using cache
- name: Build runtime image ${{ matrix.base_image.image }} for fork
if: github.event.pull_request.head.repo.fork
uses: useblacksmith/build-push-action@v1
with:
load: true
tags: ghcr.io/${{ env.REPO_OWNER }}/runtime:${{ env.RELEVANT_SHA }}-${{ matrix.base_image.tag }}
context: containers/runtime
- name: Install poetry via pipx
run: pipx install poetry
- name: Set up Python
uses: useblacksmith/setup-python@v6
with:
python-version: "3.12"
cache: poetry
- name: Install Python dependencies using Poetry
run: make install-python-dependencies INSTALL_PLAYWRIGHT=0
- name: Run docker runtime tests
shell: bash
run: |
# We install pytest-xdist in order to run tests across CPUs
poetry run pip install pytest-xdist
# Install to be able to retry on failures for flakey tests
poetry run pip install pytest-rerunfailures
image_name=ghcr.io/${{ env.REPO_OWNER }}/runtime:${{ env.RELEVANT_SHA }}-${{ matrix.base_image.tag }}
# Setting RUN_AS_OPENHANDS to false means use root.
# That should mean SANDBOX_USER_ID is ignored but some tests do not check for RUN_AS_OPENHANDS.
TEST_RUNTIME=docker \
SANDBOX_USER_ID=$(id -u) \
SANDBOX_RUNTIME_CONTAINER_IMAGE=$image_name \
TEST_IN_CI=true \
RUN_AS_OPENHANDS=false \
poetry run pytest -n 5 -raRs --reruns 2 --reruns-delay 3 -s ./tests/runtime --ignore=tests/runtime/test_browsergym_envs.py --durations=10
env:
DEBUG: "1"
# Run unit tests with the Docker runtime Docker images as openhands user
test_runtime_oh:
name: RT Unit Tests (openhands)
runs-on: blacksmith-4vcpu-ubuntu-2404
needs: [ghcr_build_runtime, define-matrix]
strategy:
matrix:
base_image: ${{ fromJson(needs.define-matrix.outputs.base_image) }}
steps:
- uses: actions/checkout@v4
- name: Set up Docker Buildx
id: buildx
uses: docker/setup-buildx-action@v3
- name: Download runtime source for fork
if: github.event.pull_request.head.repo.fork
uses: actions/download-artifact@v6
with:
name: runtime-src-${{ matrix.base_image.tag }}
path: containers/runtime
- name: Lowercase Repository Owner
run: |
echo REPO_OWNER=$(echo ${{ github.repository_owner }} | tr '[:upper:]' '[:lower:]') >> $GITHUB_ENV
# Forked repos can't push to GHCR, so we need to rebuild using cache
- name: Build runtime image ${{ matrix.base_image.image }} for fork
if: github.event.pull_request.head.repo.fork
uses: useblacksmith/build-push-action@v1
with:
load: true
tags: ghcr.io/${{ env.REPO_OWNER }}/runtime:${{ env.RELEVANT_SHA }}-${{ matrix.base_image.tag }}
context: containers/runtime
- name: Install poetry via pipx
run: pipx install poetry
- name: Set up Python
uses: useblacksmith/setup-python@v6
with:
python-version: "3.12"
cache: poetry
- name: Install Python dependencies using Poetry
run: make install-python-dependencies POETRY_GROUP=main,test,runtime INSTALL_PLAYWRIGHT=0
- name: Run runtime tests
shell: bash
run: |
# We install pytest-xdist in order to run tests across CPUs
poetry run pip install pytest-xdist
# Install to be able to retry on failures for flaky tests
poetry run pip install pytest-rerunfailures
image_name=ghcr.io/${{ env.REPO_OWNER }}/runtime:${{ env.RELEVANT_SHA }}-${{ matrix.base_image.tag }}
TEST_RUNTIME=docker \
SANDBOX_USER_ID=$(id -u) \
SANDBOX_RUNTIME_CONTAINER_IMAGE=$image_name \
TEST_IN_CI=true \
RUN_AS_OPENHANDS=true \
poetry run pytest -n 5 -raRs --reruns 2 --reruns-delay 3 -s ./tests/runtime --ignore=tests/runtime/test_browsergym_envs.py --durations=10
env:
DEBUG: "1"
# The two following jobs (named identically) are to check whether all the runtime tests have passed as the
# "All Runtime Tests Passed" is a required job for PRs to merge
# Due to this bug: https://github.com/actions/runner/issues/2566, we want to create a job that runs when the
# prerequisites have been cancelled or failed so merging is disallowed, otherwise Github considers "skipped" as "success"
# We can remove this once the config changes
runtime_tests_check_success:
name: All Runtime Tests Passed
if: ${{ !cancelled() && !contains(needs.*.result, 'failure') && !contains(needs.*.result, 'cancelled') }}
runs-on: blacksmith-4vcpu-ubuntu-2204
needs: [test_runtime_root, test_runtime_oh]
steps:
- name: All tests passed
run: echo "All runtime tests have passed successfully!"
runtime_tests_check_fail:
name: All Runtime Tests Passed
if: ${{ cancelled() || contains(needs.*.result, 'failure') || contains(needs.*.result, 'cancelled') }}
runs-on: blacksmith-4vcpu-ubuntu-2204
needs: [test_runtime_root, test_runtime_oh]
steps:
- name: Some tests failed
run: |
echo "Some runtime tests failed or were cancelled"
exit 1
update_pr_description:
name: Update PR Description
if: github.event_name == 'pull_request' && !github.event.pull_request.head.repo.fork && github.actor != 'dependabot[bot]'

View File

@@ -118,7 +118,7 @@ jobs:
contains(github.event.review.body, '@openhands-agent-exp')
)
)
uses: actions/cache@v4
uses: actions/cache@v5
with:
path: ${{ env.pythonLocation }}/lib/python3.12/site-packages/*
key: ${{ runner.os }}-pip-openhands-resolver-${{ hashFiles('/tmp/requirements.txt') }}

View File

@@ -0,0 +1,48 @@
---
name: PR Review by OpenHands
on:
# TEMPORARY MITIGATION (Clinejection hardening)
#
# We temporarily avoid `pull_request_target` here. We'll restore it after the PR review
# workflow is fully hardened for untrusted execution.
pull_request:
types: [opened, ready_for_review, labeled, review_requested]
permissions:
contents: read
pull-requests: write
issues: write
jobs:
pr-review:
# Note: fork PRs will not have access to repository secrets under `pull_request`.
# Skip forks to avoid noisy failures until we restore a hardened `pull_request_target` flow.
if: |
github.event.pull_request.head.repo.full_name == github.repository &&
(
(github.event.action == 'opened' && github.event.pull_request.draft == false) ||
github.event.action == 'ready_for_review' ||
(github.event.action == 'labeled' && github.event.label.name == 'review-this') ||
(
github.event.action == 'review_requested' &&
(
github.event.requested_reviewer.login == 'openhands-agent' ||
github.event.requested_reviewer.login == 'all-hands-bot'
)
)
)
concurrency:
group: pr-review-${{ github.event.pull_request.number }}
cancel-in-progress: true
runs-on: ubuntu-24.04
steps:
- name: Run PR Review
uses: OpenHands/extensions/plugins/pr-review@main
with:
llm-model: litellm_proxy/claude-sonnet-4-5-20250929
llm-base-url: https://llm-proxy.app.all-hands.dev
review-style: roasted
llm-api-key: ${{ secrets.LLM_API_KEY }}
github-token: ${{ secrets.ALLHANDS_BOT_GITHUB_PAT }}
lmnr-api-key: ${{ secrets.LMNR_SKILLS_API_KEY }}

View File

@@ -0,0 +1,85 @@
---
name: PR Review Evaluation
# This workflow evaluates how well PR review comments were addressed.
# It runs when a PR is closed to assess review effectiveness.
#
# Security note: pull_request_target is safe here because:
# 1. Only triggers on PR close (not on code changes)
# 2. Does not checkout PR code - only downloads artifacts from trusted workflow runs
# 3. Runs evaluation scripts from the extensions repo, not from the PR
on:
pull_request_target:
types: [closed]
permissions:
contents: read
pull-requests: read
jobs:
evaluate:
runs-on: ubuntu-24.04
env:
PR_NUMBER: ${{ github.event.pull_request.number }}
REPO_NAME: ${{ github.repository }}
PR_MERGED: ${{ github.event.pull_request.merged }}
steps:
- name: Download review trace artifact
id: download-trace
uses: dawidd6/action-download-artifact@v6
continue-on-error: true
with:
workflow: pr-review-by-openhands.yml
name: pr-review-trace-${{ github.event.pull_request.number }}
path: trace-info
search_artifacts: true
if_no_artifact_found: warn
- name: Check if trace file exists
id: check-trace
run: |
if [ -f "trace-info/laminar_trace_info.json" ]; then
echo "trace_exists=true" >> $GITHUB_OUTPUT
echo "Found trace file for PR #$PR_NUMBER"
else
echo "trace_exists=false" >> $GITHUB_OUTPUT
echo "No trace file found for PR #$PR_NUMBER - skipping evaluation"
fi
# Always checkout main branch for security - cannot test script changes in PRs
- name: Checkout extensions repository
if: steps.check-trace.outputs.trace_exists == 'true'
uses: actions/checkout@v5
with:
repository: OpenHands/extensions
path: extensions
- name: Set up Python
if: steps.check-trace.outputs.trace_exists == 'true'
uses: actions/setup-python@v6
with:
python-version: '3.12'
- name: Install dependencies
if: steps.check-trace.outputs.trace_exists == 'true'
run: pip install lmnr
- name: Run evaluation
if: steps.check-trace.outputs.trace_exists == 'true'
env:
# Script expects LMNR_PROJECT_API_KEY; org secret is named LMNR_SKILLS_API_KEY
LMNR_PROJECT_API_KEY: ${{ secrets.LMNR_SKILLS_API_KEY }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
python extensions/plugins/pr-review/scripts/evaluate_review.py \
--trace-file trace-info/laminar_trace_info.json
- name: Upload evaluation logs
uses: actions/upload-artifact@v5
if: always() && steps.check-trace.outputs.trace_exists == 'true'
with:
name: pr-review-evaluation-${{ github.event.pull_request.number }}
path: '*.log'
retention-days: 30

View File

@@ -9,6 +9,7 @@ on:
jobs:
stale:
runs-on: blacksmith-4vcpu-ubuntu-2204
if: github.repository == 'OpenHands/OpenHands'
steps:
- uses: actions/stale@v9
with:

View File

@@ -1,156 +0,0 @@
# Workflow that validates the VSCode extension builds correctly
name: VSCode Extension CI
# * Always run on "main"
# * Run on PRs that have changes in the VSCode extension folder or this workflow
# * Run on tags that start with "ext-v"
on:
push:
branches:
- main
tags:
- 'ext-v*'
pull_request:
paths:
- 'openhands/integrations/vscode/**'
- 'build_vscode.py'
- '.github/workflows/vscode-extension-build.yml'
# If triggered by a PR, it will be in the same group. However, each commit on main will be in its own unique group
concurrency:
group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
cancel-in-progress: true
jobs:
# Validate VSCode extension builds correctly
validate-vscode-extension:
name: Validate VSCode Extension Build
runs-on: blacksmith-4vcpu-ubuntu-2204
steps:
- name: Checkout
uses: actions/checkout@v4
- name: Set up Node.js
uses: useblacksmith/setup-node@v5
with:
node-version: '22'
- name: Set up Python
uses: actions/setup-python@v6
with:
python-version: '3.12'
- name: Install VSCode extension dependencies
working-directory: ./openhands/integrations/vscode
run: npm ci
- name: Build VSCode extension via build_vscode.py
run: python build_vscode.py
env:
# Ensure we don't skip the build
SKIP_VSCODE_BUILD: ""
- name: Validate .vsix file
run: |
# Verify the .vsix was created and is valid
if [ -f "openhands/integrations/vscode/openhands-vscode-0.0.1.vsix" ]; then
echo "✅ VSCode extension built successfully"
ls -la openhands/integrations/vscode/openhands-vscode-0.0.1.vsix
# Basic validation that the .vsix is a valid zip file
echo "🔍 Validating .vsix structure..."
file openhands/integrations/vscode/openhands-vscode-0.0.1.vsix
unzip -t openhands/integrations/vscode/openhands-vscode-0.0.1.vsix
echo "✅ VSCode extension validation passed"
else
echo "❌ VSCode extension build failed - .vsix not found"
exit 1
fi
- name: Upload VSCode extension artifact
uses: actions/upload-artifact@v6
with:
name: vscode-extension
path: openhands/integrations/vscode/openhands-vscode-0.0.1.vsix
retention-days: 7
- name: Comment on PR with artifact link
if: github.event_name == 'pull_request'
uses: actions/github-script@v7
with:
script: |
const fs = require('fs');
const path = require('path');
// Get file size for display
const vsixPath = 'openhands/integrations/vscode/openhands-vscode-0.0.1.vsix';
const stats = fs.statSync(vsixPath);
const fileSizeKB = Math.round(stats.size / 1024);
const comment = `## 🔧 VSCode Extension Built Successfully!
The VSCode extension has been built and is ready for testing.
**📦 Download**: [openhands-vscode-0.0.1.vsix](https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}) (${fileSizeKB} KB)
**🚀 To install**:
1. Download the artifact from the workflow run above
2. In VSCode: \`Ctrl+Shift+P\` → "Extensions: Install from VSIX..."
3. Select the downloaded \`.vsix\` file
**✅ Tested with**: Node.js 22
**🔍 Validation**: File structure and integrity verified
---
*Built from commit ${{ github.sha }}*`;
// Check if we already commented on this PR and delete it
const { data: comments } = await github.rest.issues.listComments({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
});
const botComment = comments.find(comment =>
comment.user.login === 'github-actions[bot]' &&
comment.body.includes('VSCode Extension Built Successfully')
);
if (botComment) {
await github.rest.issues.deleteComment({
owner: context.repo.owner,
repo: context.repo.repo,
comment_id: botComment.id,
});
}
// Create a new comment
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
body: comment
});
release:
name: Create GitHub Release
runs-on: blacksmith-4vcpu-ubuntu-2204
needs: validate-vscode-extension
if: startsWith(github.ref, 'refs/tags/ext-v')
steps:
- name: Download .vsix artifact
uses: actions/download-artifact@v6
with:
name: vscode-extension
path: ./
- name: Create Release
uses: ncipollo/release-action@v1.16.0
with:
artifacts: "*.vsix"
token: ${{ secrets.GITHUB_TOKEN }}
draft: true
allowUpdates: true

View File

@@ -45,7 +45,7 @@ jobs:
"This issue has been labeled as **good first issue**, which means it's a great place to get started with the OpenHands project.\n\n" +
"If you're interested in working on it, feel free to! No need to ask for permission.\n\n" +
"Be sure to check out our [development setup guide](" + repoUrl + "/blob/main/Development.md) to get your environment set up, and follow our [contribution guidelines](" + repoUrl + "/blob/main/CONTRIBUTING.md) when you're ready to submit a fix.\n\n" +
"Feel free to join our developer community on [Slack](https://all-hands.dev/joinslack). You can ask for [help](https://openhands-ai.slack.com/archives/C078L0FUGUX), [feedback](https://openhands-ai.slack.com/archives/C086ARSNMGA), and even ask for a [PR review](https://openhands-ai.slack.com/archives/C08D8FJ5771).\n\n" +
"Feel free to join our developer community on [Slack](https://openhands.dev/joinslack). You can ask for [help](https://openhands-ai.slack.com/archives/C078L0FUGUX), [feedback](https://openhands-ai.slack.com/archives/C086ARSNMGA), and even ask for a [PR review](https://openhands-ai.slack.com/archives/C08D8FJ5771).\n\n" +
"🙌 Happy hacking! 🙌\n\n" +
"<!-- auto-comment:good-first-issue -->"
});

2
.gitignore vendored
View File

@@ -234,6 +234,8 @@ yarn-error.log*
logs
ralph/
# agent
.envrc
/workspace

View File

@@ -13,7 +13,6 @@ STAGED_FILES=$(git diff --cached --name-only)
# Check if any files match specific patterns
has_frontend_changes=false
has_backend_changes=false
has_vscode_changes=false
# Check each file individually to avoid issues with grep
for file in $STAGED_FILES; do
@@ -21,17 +20,12 @@ for file in $STAGED_FILES; do
has_frontend_changes=true
elif [[ $file == openhands/* || $file == evaluation/* || $file == tests/* ]]; then
has_backend_changes=true
# Check for VSCode extension changes (subset of backend changes)
if [[ $file == openhands/integrations/vscode/* ]]; then
has_vscode_changes=true
fi
fi
done
echo "Analyzing changes..."
echo "- Frontend changes: $has_frontend_changes"
echo "- Backend changes: $has_backend_changes"
echo "- VSCode extension changes: $has_vscode_changes"
# Run frontend linting if needed
if [ "$has_frontend_changes" = true ]; then
@@ -92,51 +86,6 @@ else
echo "Skipping backend checks (no backend changes detected)."
fi
# Run VSCode extension checks if needed
if [ "$has_vscode_changes" = true ]; then
# Check if we're in a CI environment
if [ -n "$CI" ]; then
echo "Skipping VSCode extension checks (CI environment detected)."
echo "WARNING: VSCode extension files have changed but checks are being skipped."
echo "Please run VSCode extension checks manually before submitting your PR."
else
echo "Running VSCode extension checks..."
if [ -d "openhands/integrations/vscode" ]; then
cd openhands/integrations/vscode || exit 1
echo "Running npm lint:fix..."
npm run lint:fix
if [ $? -ne 0 ]; then
echo "VSCode extension linting failed. Please fix the issues before committing."
EXIT_CODE=1
else
echo "VSCode extension linting passed!"
fi
echo "Running npm typecheck..."
npm run typecheck
if [ $? -ne 0 ]; then
echo "VSCode extension type checking failed. Please fix the issues before committing."
EXIT_CODE=1
else
echo "VSCode extension type checking passed!"
fi
echo "Running npm compile..."
npm run compile
if [ $? -ne 0 ]; then
echo "VSCode extension compilation failed. Please fix the issues before committing."
EXIT_CODE=1
else
echo "VSCode extension compilation passed!"
fi
cd ../../..
fi
fi
else
echo "Skipping VSCode extension checks (no VSCode extension changes detected)."
fi
# If no specific code changes detected, run basic checks
if [ "$has_frontend_changes" = false ] && [ "$has_backend_changes" = false ]; then

View File

@@ -39,6 +39,8 @@ then re-run the command to ensure it passes. Common issues include:
## Repository Structure
Backend:
- Located in the `openhands` directory
- The current V1 application server lives in `openhands/app_server/`. `make start-backend` still launches `openhands.server.listen:app`, which includes the V1 routes by default unless `ENABLE_V1=0`.
- For V1 web-app docs, LLM setup should point users to the Settings UI.
- Testing:
- All tests are in `tests/unit/test_*.py`
- To test new code, run `poetry run pytest tests/unit/test_xxx.py` where `xxx` is the appropriate file for the current functionality
@@ -150,9 +152,9 @@ Each integration follows a consistent pattern with service classes, storage mode
**Important Notes:**
- Enterprise code is licensed under Polyform Free Trial License (30-day limit)
- The enterprise server extends the OSS server through dynamic imports
- The enterprise server extends the OpenHands server through dynamic imports
- Database changes require careful migration planning in `enterprise/migrations/`
- Always test changes in both OSS and enterprise contexts
- Always test changes in both OpenHands and enterprise contexts
- Use the enterprise-specific Makefile commands for development
**Enterprise Testing Best Practices:**
@@ -165,8 +167,8 @@ Each integration follows a consistent pattern with service classes, storage mode
**Import Patterns:**
- Use relative imports without `enterprise.` prefix in enterprise code
- Example: `from storage.database import session_maker` not `from enterprise.storage.database import session_maker`
- This ensures code works in both OSS and enterprise contexts
- Example: `from storage.database import a_session_maker` not `from enterprise.storage.database import a_session_maker`
- This ensures code works in both OpenHands and enterprise contexts
**Test Structure:**
- Place tests in `enterprise/tests/unit/` following the same structure as the source code
@@ -342,3 +344,30 @@ To add a new LLM model to OpenHands, you need to update multiple files across bo
- Models appear in CLI provider selection based on the verified arrays
- The `organize_models_and_providers` function groups models by provider
- Default model selection prioritizes verified models for each provider
### Sandbox Settings API (SDK Credential Inheritance)
The sandbox settings API allows SDK-created conversations to inherit the user's SaaS credentials
(LLM config, secrets) securely via `LookupSecret`. Raw secret values only flow SaaS→sandbox,
never through the SDK client.
#### User Credentials with Exposed Secrets (in `openhands/app_server/user/user_router.py`):
- `GET /api/v1/users/me?expose_secrets=true` → Full user settings with unmasked secrets (e.g., `llm_api_key`)
- `GET /api/v1/users/me` → Full user settings (secrets masked, Bearer only)
Auth requirements for `expose_secrets=true`:
- Bearer token (proves user identity via `OPENHANDS_API_KEY`)
- `X-Session-API-Key` header (proves caller has an active sandbox owned by the authenticated user)
Called by `workspace.get_llm()` in the SDK to retrieve LLM config with the API key.
#### Sandbox-Scoped Secrets Endpoints (in `openhands/app_server/sandbox/sandbox_router.py`):
- `GET /sandboxes/{id}/settings/secrets` → list secret names (no values)
- `GET /sandboxes/{id}/settings/secrets/{name}` → raw secret value (called FROM sandbox)
#### Auth: `X-Session-API-Key` header, validated via `SandboxService.get_sandbox_by_session_api_key()`
#### Related SDK code (in `software-agent-sdk` repo):
- `openhands/sdk/llm/llm.py`: `LLM.api_key` accepts `SecretSource` (including `LookupSecret`)
- `openhands/workspace/cloud/workspace.py`: `get_llm()` and `get_secrets()` return LookupSecret-backed objects
- Tests: `tests/sdk/llm/test_llm_secret_source_api_key.py`, `tests/workspace/test_cloud_workspace_sdk_settings.py`

View File

@@ -61,7 +61,7 @@ representative at an online or offline event.
Instances of abusive, harassing, or otherwise unacceptable behavior may be
reported to the community leaders responsible for enforcement at
contact@all-hands.dev.
contact@openhands.dev.
All complaints will be reviewed and investigated promptly and fairly.
All community leaders are obligated to respect the privacy and security of the
@@ -115,7 +115,9 @@ community.
### Slack Etiquettes
These Slack etiquette guidelines are designed to foster an inclusive, respectful, and productive environment for all community members. By following these best practices, we ensure effective communication and collaboration while minimizing disruptions. Lets work together to build a supportive and welcoming community!
These Slack etiquette guidelines are designed to foster an inclusive, respectful, and productive environment for all
community members. By following these best practices, we ensure effective communication and collaboration while
minimizing disruptions. Lets work together to build a supportive and welcoming community!
- Communicate respectfully and professionally, avoiding sarcasm or harsh language, and remember that tone can be difficult to interpret in text.
- Use threads for specific discussions to keep channels organized and easier to follow.
@@ -125,7 +127,10 @@ These Slack etiquette guidelines are designed to foster an inclusive, respectful
- When asking for help or raising issues, include necessary details like links, screenshots, or clear explanations to provide context.
- Keep discussions in public channels whenever possible to allow others to benefit from the conversation, unless the matter is sensitive or private.
- Always adhere to [our standards](https://github.com/OpenHands/OpenHands/blob/main/CODE_OF_CONDUCT.md#our-standards) to ensure a welcoming and collaborative environment.
- If you choose to mute a channel, consider setting up alerts for topics that still interest you to stay engaged. For Slack, Go to Settings → Notifications → My Keywords to add specific keywords that will notify you when mentioned. For example, if you're here for discussions about LLMs, mute the channel if its too busy, but set notifications to alert you only when “LLMs” appears in messages.
- If you choose to mute a channel, consider setting up alerts for topics that still interest you to stay engaged.
For Slack, Go to Settings → Notifications → My Keywords to add specific keywords that will notify you when mentioned.
For example, if you're here for discussions about LLMs, mute the channel if its too busy, but set notifications to
alert you only when “LLMs” appears in messages.
## Attribution

View File

@@ -1,36 +1,44 @@
# The OpenHands Community
OpenHands is a community of engineers, academics, and enthusiasts reimagining software development for an AI-powered world.
OpenHands is a community of engineers, academics, and enthusiasts reimagining software development for an AI-powered
world.
## Mission
Its very clear that AI is changing software development. We want the developer community to drive that change organically, through open source.
Its very clear that AI is changing software development. We want the developer community to drive that change
organically, through open source.
So were not just building friendly interfaces for AI-driven development. Were publishing _building blocks_ that empower developers to create new experiences, tailored to your own habits, needs, and imagination.
So were not just building friendly interfaces for AI-driven development. Were publishing _building blocks_ that
empower developers to create new experiences, tailored to your own habits, needs, and imagination.
## Ethos
We have two core values: **high openness** and **high agency**. While we dont expect everyone in the community to embody these values, we want to establish them as norms.
We have two core values: **high openness** and **high agency**. While we dont expect everyone in the community to
embody these values, we want to establish them as norms.
### High Openness
We welcome anyone and everyone into our community by default. You dont have to be a software developer to help us build. You dont have to be pro-AI to help us learn.
We welcome anyone and everyone into our community by default. You dont have to be a software developer to help us
build. You dont have to be pro-AI to help us learn.
Our plans, our work, our successes, and our failures are all public record. We want the world to see not just the fruits of our work, but the whole process of growing it.
Our plans, our work, our successes, and our failures are all public record. We want the world to see not just the
fruits of our work, but the whole process of growing it.
We welcome thoughtful criticism, whether its a comment on a PR or feedback on the community as a whole.
### High Agency
Everyone should feel empowered to contribute to OpenHands. Whether its by making a PR, hosting an event, sharing feedback, or just asking a question, dont hold back!
Everyone should feel empowered to contribute to OpenHands. Whether its by making a PR, hosting an event, sharing
feedback, or just asking a question, dont hold back!
OpenHands gives everyone the building blocks to create state-of-the-art developer experiences. We experiment constantly and love building new things.
OpenHands gives everyone the building blocks to create state-of-the-art developer experiences. We experiment constantly
and love building new things.
Coding, development practices, and communities are changing rapidly. We wont hesitate to change direction and make big bets.
## Relationship to All Hands
OpenHands is supported by the for-profit organization [All Hands AI, Inc](https://www.all-hands.dev/).
OpenHands is supported by the for-profit organization [All Hands AI, Inc](https://www.openhands.dev/).
All Hands was founded by three of the first major contributors to OpenHands:
@@ -38,8 +46,13 @@ All Hands was founded by three of the first major contributors to OpenHands:
- Graham Neubig, a CMU Professor who rallied the academic community around OpenHands
- Robert Brennan, a software engineer who architected the user-facing features of OpenHands
All Hands is an important part of the OpenHands ecosystem. Weve raised over $20M--mainly to hire developers and researchers who can work on OpenHands full-time, and to provide them with expensive infrastructure. ([Join us!](https://allhandsai.applytojob.com/apply/))
All Hands is an important part of the OpenHands ecosystem. Weve raised over $20M--mainly to hire developers and
researchers who can work on OpenHands full-time, and to provide them with expensive infrastructure. ([Join us!](https://allhandsai.applytojob.com/apply/))
But we see OpenHands as much larger, and ultimately more important, than All Hands. When our financial responsibility to investors is at odds with our social responsibility to the community—as it inevitably will be, from time to time—we promise to navigate that conflict thoughtfully and transparently.
But we see OpenHands as much larger, and ultimately more important, than All Hands. When our financial responsibility
to investors is at odds with our social responsibility to the community—as it inevitably will be, from time to time—we
promise to navigate that conflict thoughtfully and transparently.
At some point, we may transfer custody of OpenHands to an open source foundation. But for now, the [Benevolent Dictator approach](http://www.catb.org/~esr/writings/cathedral-bazaar/homesteading/ar01s16.html) helps us move forward with speed and intention. If we ever forget the “benevolent” part, please: fork us.
At some point, we may transfer custody of OpenHands to an open source foundation. But for now,
the [Benevolent Dictator approach](http://www.catb.org/~esr/writings/cathedral-bazaar/homesteading/ar01s16.html) helps us move forward with speed and intention. If we ever forget the
“benevolent” part, please: fork us.

View File

@@ -1,71 +1,104 @@
# Contributing
Thanks for your interest in contributing to OpenHands! We welcome and appreciate contributions.
Thanks for your interest in contributing to OpenHands! We're building the future of AI-powered software development, and we'd love for you to be part of this journey.
## Understanding OpenHands's CodeBase
## Our Vision
To understand the codebase, please refer to the README in each module:
- [frontend](./frontend/README.md)
- [evaluation](./evaluation/README.md)
- [openhands](./openhands/README.md)
- [agenthub](./openhands/agenthub/README.md)
- [server](./openhands/server/README.md)
The OpenHands community is built around the belief that AI and AI agents are going to fundamentally change the way we build software. If this is true, we should do everything we can to make sure that the benefits provided by such powerful technology are accessible to everyone.
## Setting up Your Development Environment
We believe in the power of open source to democratize access to cutting-edge AI technology. Just as the internet transformed how we share information, we envision a world where AI-powered development tools are available to every developer, regardless of their background or resources.
We have a separate doc [Development.md](https://github.com/OpenHands/OpenHands/blob/main/Development.md) that tells you how to set up a development workflow.
## Getting Started
## How Can I Contribute?
### Quick Ways to Contribute
There are many ways that you can contribute:
- **Use OpenHands** and [report issues](https://github.com/OpenHands/OpenHands/issues) you encounter
- **Give feedback** using the thumbs-up/thumbs-down buttons after each session
- **Star our repository** on [GitHub](https://github.com/OpenHands/OpenHands)
- **Share OpenHands** with other developers
1. **Download and use** OpenHands, and send [issues](https://github.com/OpenHands/OpenHands/issues) when you encounter something that isn't working or a feature that you'd like to see.
2. **Send feedback** after each session by [clicking the thumbs-up thumbs-down buttons](https://docs.all-hands.dev/usage/feedback), so we can see where things are working and failing, and also build an open dataset for training code agents.
3. **Improve the Codebase** by sending [PRs](#sending-pull-requests-to-openhands) (see details below). In particular, we have some [good first issues](https://github.com/OpenHands/OpenHands/labels/good%20first%20issue) that may be ones to start on.
### Set Up Your Development Environment
## What Can I Build?
Here are a few ways you can help improve the codebase.
- **Requirements**: Linux/Mac/WSL, Docker, Python 3.12, Node.js 22+, Poetry 1.8+
- **Quick setup**: `make build`
- **Run locally**: `make run`
- **LLM setup (V1 web app)**: configure your model and API key in the Settings UI after the app starts
#### UI/UX
We're always looking to improve the look and feel of the application. If you've got a small fix
for something that's bugging you, feel free to open up a PR that changes the [`./frontend`](./frontend) directory.
Full details in our [Development Guide](./Development.md).
If you're looking to make a bigger change, add a new UI element, or significantly alter the style
of the application, please open an issue first, or better, join the #eng-ui-ux channel in our Slack
to gather consensus from our design team first.
### Find Your First Issue
#### Improving the agent
Our main agent is the CodeAct agent. You can [see its prompts here](https://github.com/OpenHands/OpenHands/tree/main/openhands/agenthub/codeact_agent).
- Browse [good first issues](https://github.com/OpenHands/OpenHands/labels/good%20first%20issue)
- Check our [project boards](https://github.com/OpenHands/OpenHands/projects) for organized tasks
- Join our [Slack community](https://openhands.dev/joinslack) to ask what needs help
Changes to these prompts, and to the underlying behavior in Python, can have a huge impact on user experience.
You can try modifying the prompts to see how they change the behavior of the agent as you use the app
locally, but we will need to do an end-to-end evaluation of any changes here to ensure that the agent
is getting better over time.
## Understanding the Codebase
We use the [SWE-bench](https://www.swebench.com/) benchmark to test our agent. You can join the #evaluation
channel in Slack to learn more.
- **[Frontend](./frontend/README.md)** - React application
- **[App Server (V1)](./openhands/app_server/README.md)** - Current FastAPI application server and REST API modules
- **[Agents](./openhands/agenthub/README.md)** - AI agent implementations
- **[Runtime](./openhands/runtime/README.md)** - Execution environments
- **[Evaluation](https://github.com/OpenHands/benchmarks)** - Testing and benchmarks
#### Adding a new agent
You may want to experiment with building new types of agents. You can add an agent to [`openhands/agenthub`](./openhands/agenthub)
to help expand the capabilities of OpenHands.
## What Can You Build?
#### Adding a new runtime
The agent needs a place to run code and commands. When you run OpenHands on your laptop, it uses a Docker container
to do this by default. But there are other ways of creating a sandbox for the agent.
### Frontend & UI/UX
- React & TypeScript development
- UI/UX improvements
- Mobile responsiveness
- Component libraries
If you work for a company that provides a cloud-based runtime, you could help us add support for that runtime
by implementing the [interface specified here](https://github.com/OpenHands/OpenHands/blob/main/openhands/runtime/base.py).
For bigger changes, join the #proj-gui channel in [Slack](https://openhands.dev/joinslack) first.
#### Testing
When you write code, it is also good to write tests. Please navigate to the [`./tests`](./tests) folder to see existing test suites.
At the moment, we have these kinds of tests: [`unit`](./tests/unit), [`runtime`](./tests/runtime), and [`end-to-end (e2e)`](./tests/e2e). Please refer to the README for each test suite. These tests also run on GitHub's continuous integration to ensure quality of the project.
### Agent Development
- Prompt engineering
- New agent types
- Agent evaluation
- Multi-agent systems
We use [SWE-bench](https://www.swebench.com/) to evaluate agents.
### Backend & Infrastructure
- Python development
- Runtime systems (Docker containers, sandboxes)
- Cloud integrations
- Performance optimization
### Testing & Quality Assurance
- Unit testing
- Integration testing
- Bug hunting
- Performance testing
### Documentation & Education
- Technical documentation
- Translation
- Community support
## Pull Request Process
### Small Improvements
- Quick review and approval
- Ensure CI tests pass
- Include clear description of changes
### Core Agent Changes
These are evaluated based on:
- **Accuracy** - Does it make the agent better at solving problems?
- **Efficiency** - Does it improve speed or reduce resource usage?
- **Code Quality** - Is the code maintainable and well-tested?
Discuss major changes in [GitHub issues](https://github.com/OpenHands/OpenHands/issues) or [Slack](https://openhands.dev/joinslack) first.
## Sending Pull Requests to OpenHands
You'll need to fork our repository to send us a Pull Request. You can learn more
about how to fork a GitHub repo and open a PR with your changes in [this article](https://medium.com/swlh/forks-and-pull-requests-how-to-contribute-to-github-repos-8843fac34ce8).
### Pull Request title
You may also check out previous PRs in the [PR list](https://github.com/OpenHands/OpenHands/pulls).
### Pull Request Title Format
As described [here](https://github.com/commitizen/conventional-commit-types/blob/master/index.json), a valid PR title should begin with one of the following prefixes:
- `feat`: A new feature
@@ -84,40 +117,16 @@ For example, a PR title could be:
- `refactor: modify package path`
- `feat(frontend): xxxx`, where `(frontend)` means that this PR mainly focuses on the frontend component.
You may also check out previous PRs in the [PR list](https://github.com/OpenHands/OpenHands/pulls).
### Pull Request Description
### Pull Request description
- If your PR is small (such as a typo fix), you can go brief.
- If it contains a lot of changes, it's better to write more details.
- Explain what the PR does and why
- Link to related issues
- Include screenshots for UI changes
- If your changes are user-facing (e.g. a new feature in the UI, a change in behavior, or a bugfix),
please include a short message that we can add to our changelog
If your changes are user-facing (e.g. a new feature in the UI, a change in behavior, or a bugfix)
please include a short message that we can add to our changelog.
## Need Help?
## How to Make Effective Contributions
### Opening Issues
If you notice any bugs or have any feature requests please open them via the [issues page](https://github.com/OpenHands/OpenHands/issues). We will triage based on how critical the bug is or how potentially useful the improvement is, discuss, and implement the ones that the community has interest/effort for.
Further, if you see an issue you like, please leave a "thumbs-up" or a comment, which will help us prioritize.
### Making Pull Requests
We're generally happy to consider all pull requests with the evaluation process varying based on the type of change:
#### For Small Improvements
Small improvements with few downsides are typically reviewed and approved quickly.
One thing to check when making changes is to ensure that all continuous integration tests pass, which you can check before getting a review.
#### For Core Agent Changes
We need to be more careful with changes to the core agent, as it is imperative to maintain high quality. These PRs are evaluated based on three key metrics:
1. **Accuracy**
2. **Efficiency**
3. **Code Complexity**
If it improves accuracy, efficiency, or both with only a minimal change to code quality, that's great we're happy to merge it in!
If there are bigger tradeoffs (e.g. helping efficiency a lot and hurting accuracy a little) we might want to put it behind a feature flag.
Either way, please feel free to discuss on github issues or slack, and we will give guidance and preliminary feedback.
- **Slack**: [Join our community](https://openhands.dev/joinslack)
- **GitHub Issues**: [Open an issue](https://github.com/OpenHands/OpenHands/issues)
- **Email**: contact@openhands.dev

View File

@@ -2,11 +2,13 @@
## Contributors
We would like to thank all the [contributors](https://github.com/OpenHands/OpenHands/graphs/contributors) who have helped make OpenHands possible. We greatly appreciate your dedication and hard work.
We would like to thank all the [contributors](https://github.com/OpenHands/OpenHands/graphs/contributors) who have
helped make OpenHands possible. We greatly appreciate your dedication and hard work.
## Open Source Projects
OpenHands includes and adapts the following open source projects. We are grateful for their contributions to the open source community:
OpenHands includes and adapts the following open source projects. We are grateful for their contributions to the
open source community:
#### [SWE Agent](https://github.com/princeton-nlp/swe-agent)
- License: MIT License
@@ -20,8 +22,8 @@ OpenHands includes and adapts the following open source projects. We are gratefu
- License: Apache License 2.0
- Description: Adapted in implementing the browsing agent
### Reference Implementations for Evaluation Benchmarks
OpenHands integrates code of the reference implementations for the following agent evaluation benchmarks:
#### [HumanEval](https://github.com/openai/human-eval)
@@ -52,28 +54,44 @@ OpenHands integrates code of the reference implementations for the following age
#### [ProntoQA](https://github.com/asaparov/prontoqa)
- License: Apache License 2.0
## Open Source licenses
### MIT License
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
documentation files (the "Software"), to deal in the Software without restriction, including without limitation the
rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit
persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the
Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
### BSD 3-Clause License
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the
following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following
disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following
disclaimer in the documentation and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote
products derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
### Apache License 2.0
@@ -268,8 +286,6 @@ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
Copyright [yyyy] [name of copyright owner]
### Non-Open Source Reference Implementations:
#### [MultiPL-E](https://github.com/nuprl/MultiPL-E)

View File

@@ -6,22 +6,196 @@ If you wish to contribute your changes, check out the
on how to clone and setup the project initially before moving on. Otherwise,
you can clone the OpenHands project directly.
## Start the Server for Development
## Choose Your Setup
### 1. Requirements
Select your operating system to see the specific setup instructions:
- Linux, Mac OS, or [WSL on Windows](https://learn.microsoft.com/en-us/windows/wsl/install) [Ubuntu >= 22.04]
- [Docker](https://docs.docker.com/engine/install/) (For those on MacOS, make sure to allow the default Docker socket to be used from advanced settings!)
- [Python](https://www.python.org/downloads/) = 3.12
- [NodeJS](https://nodejs.org/en/download/package-manager) >= 22.x
- [Poetry](https://python-poetry.org/docs/#installing-with-the-official-installer) >= 1.8
- OS-specific dependencies:
- Ubuntu: build-essential => `sudo apt-get install build-essential python3.12-dev`
- WSL: netcat => `sudo apt-get install netcat`
- [macOS](#macos-setup)
- [Linux](#linux-setup)
- [Windows WSL](#windows-wsl-setup)
- [Dev Container](#dev-container)
- [Developing in Docker](#developing-in-docker)
- [No sudo access?](#develop-without-sudo-access)
Make sure you have all these dependencies installed before moving on to `make build`.
---
#### Dev container
## macOS Setup
### 1. Install Prerequisites
You'll need the following installed:
- **Python 3.12** — `brew install python@3.12` (see the [official Homebrew Python docs](https://docs.brew.sh/Homebrew-and-Python) for details). Make sure `python3.12` is available in your PATH (the `make build` step will verify this).
- **Node.js >= 22** — `brew install node`
- **Poetry >= 1.8** — `brew install poetry`
- **Docker Desktop** — `brew install --cask docker`
- After installing, open Docker Desktop → **Settings → Advanced** → Enable **"Allow the default Docker socket to be used"**
### 2. Build and Setup the Environment
```bash
make build
```
### 3. Configure the Language Model
OpenHands supports a diverse array of Language Models (LMs) through the powerful [litellm](https://docs.litellm.ai) library.
For the V1 web app, start OpenHands and configure your model and API key in the Settings UI.
If you are running headless or CLI workflows, you can prepare local defaults with:
```bash
make setup-config
```
**Note on Alternative Models:**
See [our documentation](https://docs.openhands.dev/usage/llms) for recommended models.
### 4. Run the Application
```bash
# Run both backend and frontend
make run
# Or run separately:
make start-backend # Backend only on port 3000
make start-frontend # Frontend only on port 3001
```
These targets serve the current OpenHands V1 API by default. In the codebase, `make start-backend` runs `openhands.server.listen:app`, and that app includes the `openhands/app_server` V1 routes unless `ENABLE_V1=0`.
---
## Linux Setup
This guide covers Ubuntu/Debian. For other distributions, adapt the package manager commands accordingly.
### 1. Install Prerequisites
```bash
# Update package list
sudo apt update
# Install system dependencies
sudo apt install -y build-essential curl netcat software-properties-common
# Install Python 3.12
# Ubuntu 24.04+ and Debian 13+ ship with Python 3.12 — skip the PPA step if
# python3.12 --version already works on your system.
# The deadsnakes PPA is Ubuntu-only and needed for Ubuntu 22.04 or older:
sudo add-apt-repository -y ppa:deadsnakes/ppa
sudo apt update
sudo apt install -y python3.12 python3.12-dev python3.12-venv
# Install Node.js 22.x
curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash -
sudo apt install -y nodejs
# Install Poetry
curl -sSL https://install.python-poetry.org | python3 -
# Add Poetry to your PATH
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc
# Install Docker
# Follow the official guide: https://docs.docker.com/engine/install/ubuntu/
# Quick version:
sudo install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc
echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt update
sudo apt install -y docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin
sudo usermod -aG docker $USER
# Log out and back in for Docker group changes to take effect
```
### 2. Build and Setup the Environment
```bash
make build
```
### 3. Configure the Language Model
See the [macOS section above](#3-configure-the-language-model) for guidance: configure your model and API key in the Settings UI.
### 4. Run the Application
```bash
# Run both backend and frontend
make run
# Or run separately:
make start-backend # Backend only on port 3000
make start-frontend # Frontend only on port 3001
```
---
## Windows WSL Setup
WSL2 with Ubuntu is recommended. The setup is similar to Linux, with a few WSL-specific considerations.
### 1. Install WSL2
**Option A: Windows 11 (Microsoft Store)**
The easiest way on Windows 11:
1. Open the **Microsoft Store** app
2. Search for **"Ubuntu 22.04 LTS"** or **"Ubuntu"**
3. Click **Install**
4. Launch Ubuntu from the Start menu
**Option B: PowerShell**
```powershell
# Run this in PowerShell as Administrator
wsl --install -d Ubuntu-22.04
```
After installation, restart your computer and open Ubuntu.
### 2. Install Prerequisites (in WSL Ubuntu)
Follow [Step 1 from the Linux setup](#1-install-prerequisites-1) to install system dependencies, Python 3.12, Node.js, and Poetry. Skip the Docker installation — Docker is provided through Docker Desktop below.
### 3. Configure Docker for WSL2
1. Install [Docker Desktop for Windows](https://www.docker.com/products/docker-desktop)
2. Open Docker Desktop > Settings > General
3. Enable: "Use the WSL 2 based engine"
4. Go to Settings > Resources > WSL Integration
5. Enable integration with your Ubuntu distribution
**Important:** Keep your project files in the WSL filesystem (e.g., `~/workspace/openhands`), not in `/mnt/c`. Files accessed via `/mnt/c` will be significantly slower.
### 4. Build and Setup the Environment
```bash
make build
```
### 5. Configure the Language Model
See the [macOS section above](#3-configure-the-language-model) for the current V1 guidance: configure your model and API key in the Settings UI for the web app, and use `make setup-config` only for headless or CLI workflows.
### 6. Run the Application
```bash
# Run both backend and frontend
make run
# Or run separately:
make start-backend # Backend only on port 3000
make start-frontend # Frontend only on port 3001
```
Access the frontend at `http://localhost:3001` from your Windows browser.
---
## Dev Container
There is a [dev container](https://containers.dev/) available which provides a
pre-configured environment with all the necessary dependencies installed if you
@@ -32,7 +206,38 @@ extension installed, you can open the project in a dev container by using the
_Dev Container: Reopen in Container_ command from the Command Palette
(Ctrl+Shift+P).
#### Develop without sudo access
---
## Developing in Docker
If you don't want to install dependencies on your host machine, you can develop inside a Docker container.
### Quick Start
```bash
make docker-dev
```
For more details, see the [dev container documentation](./containers/dev/README.md).
### Alternative: Docker Run
If you just want to run OpenHands without setting up a dev environment:
```bash
make docker-run
```
If you don't have `make` installed, run:
```bash
cd ./containers/dev
./dev.sh
```
---
## Develop without sudo access
If you want to develop without system admin/sudo access to upgrade/install `Python` and/or `NodeJS`, you can use
`conda` or `mamba` to manage the packages for you:
@@ -48,159 +253,90 @@ mamba install conda-forge::nodejs
mamba install conda-forge::poetry
```
### 2. Build and Setup The Environment
---
Begin by building the project which includes setting up the environment and installing dependencies. This step ensures
that OpenHands is ready to run on your system:
## Running OpenHands with OpenHands
You can use OpenHands to develop and improve OpenHands itself!
### Quick Start
```bash
make build
export INSTALL_DOCKER=0
export RUNTIME=local
make build && make run
```
### 3. Configuring the Language Model
Access the interface at:
- Local development: http://localhost:3001
- Remote/cloud environments: Use the appropriate external URL
OpenHands supports a diverse array of Language Models (LMs) through the powerful [litellm](https://docs.litellm.ai) library.
For external access:
```bash
make run FRONTEND_PORT=12000 FRONTEND_HOST=0.0.0.0 BACKEND_HOST=0.0.0.0
```
To configure the LM of your choice, run:
---
## LLM Debugging
If you encounter issues with the Language Model, enable debug logging:
```bash
make setup-config
export DEBUG=1
# Restart the backend
make start-backend
```
This command will prompt you to enter the LLM API key, model name, and other variables ensuring that OpenHands is
tailored to your specific needs. Note that the model name will apply only when you run headless. If you use the UI,
please set the model in the UI.
Logs will be saved to `logs/llm/CURRENT_DATE/` for troubleshooting.
Note: If you have previously run OpenHands using the docker command, you may have already set some environment
variables in your terminal. The final configurations are set from highest to lowest priority:
Environment variables > config.toml variables > default variables
---
**Note on Alternative Models:**
See [our documentation](https://docs.all-hands.dev/usage/llms) for recommended models.
## Testing
### 4. Running the application
#### Option A: Run the Full Application
Once the setup is complete, this command starts both the backend and frontend servers, allowing you to interact with OpenHands:
```bash
make run
```
#### Option B: Individual Server Startup
- **Start the Backend Server:** If you prefer, you can start the backend server independently to focus on
backend-related tasks or configurations.
```bash
make start-backend
```
- **Start the Frontend Server:** Similarly, you can start the frontend server on its own to work on frontend-related
components or interface enhancements.
```bash
make start-frontend
```
### 5. Running OpenHands with OpenHands
You can use OpenHands to develop and improve OpenHands itself! This is a powerful way to leverage AI assistance for contributing to the project.
#### Quick Start
1. **Build and run OpenHands:**
```bash
export INSTALL_DOCKER=0
export RUNTIME=local
make build && make run
```
2. **Access the interface:**
- Local development: http://localhost:3001
- Remote/cloud environments: Use the appropriate external URL
3. **Configure for external access (if needed):**
```bash
# For external access (e.g., cloud environments)
make run FRONTEND_PORT=12000 FRONTEND_HOST=0.0.0.0 BACKEND_HOST=0.0.0.0
```
### 6. LLM Debugging
If you encounter any issues with the Language Model (LM) or you're simply curious, export DEBUG=1 in the environment and restart the backend.
OpenHands will log the prompts and responses in the logs/llm/CURRENT_DATE directory, allowing you to identify the causes.
### 7. Help
Need help or info on available targets and commands? Use the help command for all the guidance you need with OpenHands.
```bash
make help
```
### 8. Testing
To run tests, refer to the following:
#### Unit tests
### Unit Tests
```bash
poetry run pytest ./tests/unit/test_*.py
```
### 9. Add or update dependency
---
1. Add your dependency in `pyproject.toml` or use `poetry add xxx`.
2. Update the poetry.lock file via `poetry lock --no-update`.
## Adding Dependencies
### 10. Use existing Docker image
1. Add your dependency in `pyproject.toml` or use `poetry add xxx`
2. Update the lock file: `poetry lock --no-update`
To reduce build time (e.g., if no changes were made to the client-runtime component), you can use an existing Docker
container image by setting the SANDBOX_RUNTIME_CONTAINER_IMAGE environment variable to the desired Docker image.
---
Example: `export SANDBOX_RUNTIME_CONTAINER_IMAGE=ghcr.io/openhands/runtime:1.0-nikolaik`
## Using Existing Docker Images
## Develop inside Docker container
TL;DR
To reduce build time, you can use an existing runtime image:
```bash
make docker-dev
export SANDBOX_RUNTIME_CONTAINER_IMAGE=ghcr.io/openhands/runtime:1.2-nikolaik
```
See more details [here](./containers/dev/README.md).
---
If you are just interested in running `OpenHands` without installing all the required tools on your host.
## Help
```bash
make docker-run
make help
```
If you do not have `make` on your host, run:
```bash
cd ./containers/dev
./dev.sh
```
You do need [Docker](https://docs.docker.com/engine/install/) installed on your host though.
---
## Key Documentation Resources
Here's a guide to the important documentation files in the repository:
- [/README.md](./README.md): Main project overview, features, and basic setup instructions
- [/Development.md](./Development.md) (this file): Comprehensive guide for developers working on OpenHands
- [/CONTRIBUTING.md](./CONTRIBUTING.md): Guidelines for contributing to the project, including code style and PR process
- [DOC_STYLE_GUIDE.md](https://github.com/All-Hands-AI/docs/blob/main/openhands/DOC_STYLE_GUIDE.md): Standards for writing and maintaining project documentation
- [/openhands/README.md](./openhands/README.md): Details about the backend Python implementation
- [DOC_STYLE_GUIDE.md](https://github.com/OpenHands/docs/blob/main/openhands/DOC_STYLE_GUIDE.md): Standards for writing and maintaining project documentation
- [/openhands/app_server/README.md](./openhands/app_server/README.md): Current V1 application server implementation and REST API modules
- [/frontend/README.md](./frontend/README.md): Frontend React application setup and development guide
- [/containers/README.md](./containers/README.md): Information about Docker containers and deployment
- [/tests/unit/README.md](./tests/unit/README.md): Guide to writing and running unit tests
- [/evaluation/README.md](./evaluation/README.md): Documentation for the evaluation framework and benchmarks
- [OpenHands/benchmarks](https://github.com/OpenHands/benchmarks): Documentation for the evaluation framework and benchmarks
- [/skills/README.md](./skills/README.md): Information about the skills architecture and implementation
- [/openhands/server/README.md](./openhands/server/README.md): Server implementation details and API documentation
- [/openhands/runtime/README.md](./openhands/runtime/README.md): Documentation for the runtime environment and execution model

View File

@@ -3,14 +3,14 @@ These are the procedures and guidelines on how issues are triaged in this repo b
## General
* All issues must be tagged with **enhancement**, **bug** or **troubleshooting/help**.
* Issues may be tagged with what it relates to (**agent quality**, **resolver**, **CLI**, etc.).
* Issues may be tagged with what it relates to (**llm**, **app tab**, **UI/UX**, etc.).
## Severity
* **High**: High visibility issues or affecting many users.
* **Critical**: Affecting all users or potential security issues.
## Difficulty
* Issues with low implementation difficulty may be tagged with **good first issue**.
* Issues good for newcomers may be tagged with **good first issue**.
## Not Enough Information
* User is asked to provide more information (logs, how to reproduce, etc.) when the issue is not clear.
@@ -22,6 +22,6 @@ the issue may be closed as **not planned** (Usually after a week).
* Issues may be broken down into multiple issues if required.
## Stale and Auto Closures
* In order to keep a maintainable backlog, issues that have no activity within 30 days are automatically marked as **Stale**.
* If issues marked as **Stale** continue to have no activity for 7 more days, they will automatically be closed as not planned.
* In order to keep a maintainable backlog, issues that have no activity within 40 days are automatically marked as **Stale**.
* If issues marked as **Stale** continue to have no activity for 10 more days, they will automatically be closed as not planned.
* Issues may be reopened by maintainers if deemed important.

View File

@@ -54,7 +54,7 @@ The experience will be familiar to anyone who has used Devin or Jules.
### OpenHands Cloud
This is a deployment of OpenHands GUI, running on hosted infrastructure.
You can try it with a free $10 credit by [signing in with your GitHub account](https://app.all-hands.dev).
You can try it for free using the Minimax model by [signing in with your GitHub or GitLab account](https://app.all-hands.dev).
OpenHands Cloud comes with source-available features and integrations:
- Integrations with Slack, Jira, and Linear

View File

@@ -1,113 +0,0 @@
import os
import pathlib
import subprocess
# This script is intended to be run by Poetry during the build process.
# Define the expected name of the .vsix file based on the extension's package.json
# This should match the name and version in openhands-vscode/package.json
EXTENSION_NAME = 'openhands-vscode'
EXTENSION_VERSION = '0.0.1'
VSIX_FILENAME = f'{EXTENSION_NAME}-{EXTENSION_VERSION}.vsix'
# Paths
ROOT_DIR = pathlib.Path(__file__).parent.resolve()
VSCODE_EXTENSION_DIR = ROOT_DIR / 'openhands' / 'integrations' / 'vscode'
def check_node_version():
"""Check if Node.js version is sufficient for building the extension."""
try:
result = subprocess.run(
['node', '--version'], capture_output=True, text=True, check=True
)
version_str = result.stdout.strip()
# Extract major version number (e.g., "v12.22.9" -> 12)
major_version = int(version_str.lstrip('v').split('.')[0])
return major_version >= 18 # Align with frontend actual usage (18.20.1)
except (subprocess.CalledProcessError, FileNotFoundError, ValueError):
return False
def build_vscode_extension():
"""Builds the VS Code extension."""
vsix_path = VSCODE_EXTENSION_DIR / VSIX_FILENAME
# Check if VSCode extension build is disabled via environment variable
if os.environ.get('SKIP_VSCODE_BUILD', '').lower() in ('1', 'true', 'yes'):
print('--- Skipping VS Code extension build (SKIP_VSCODE_BUILD is set) ---')
if vsix_path.exists():
print(f'--- Using existing VS Code extension: {vsix_path} ---')
else:
print('--- No pre-built VS Code extension found ---')
return
# Check Node.js version - if insufficient, use pre-built extension as fallback
if not check_node_version():
print('--- Warning: Node.js version < 18 detected or Node.js not found ---')
print('--- Skipping VS Code extension build (requires Node.js >= 18) ---')
print('--- Using pre-built extension if available ---')
if not vsix_path.exists():
print('--- Warning: No pre-built VS Code extension found ---')
print('--- VS Code extension will not be available ---')
else:
print(f'--- Using pre-built VS Code extension: {vsix_path} ---')
return
print(f'--- Building VS Code extension in {VSCODE_EXTENSION_DIR} ---')
try:
# Ensure npm dependencies are installed
print('--- Running npm install for VS Code extension ---')
subprocess.run(
['npm', 'install'],
cwd=VSCODE_EXTENSION_DIR,
check=True,
shell=os.name == 'nt',
)
# Package the extension
print(f'--- Packaging VS Code extension ({VSIX_FILENAME}) ---')
subprocess.run(
['npm', 'run', 'package-vsix'],
cwd=VSCODE_EXTENSION_DIR,
check=True,
shell=os.name == 'nt',
)
# Verify the generated .vsix file exists
if not vsix_path.exists():
raise FileNotFoundError(
f'VS Code extension package not found after build: {vsix_path}'
)
print(f'--- VS Code extension built successfully: {vsix_path} ---')
except subprocess.CalledProcessError as e:
print(f'--- Warning: Failed to build VS Code extension: {e} ---')
print('--- Continuing without building extension ---')
if not vsix_path.exists():
print('--- Warning: No pre-built VS Code extension found ---')
print('--- VS Code extension will not be available ---')
def build(setup_kwargs):
"""This function is called by Poetry during the build process.
`setup_kwargs` is a dictionary that will be passed to `setuptools.setup()`.
"""
print('--- Running custom Poetry build script (build_vscode.py) ---')
# Build the VS Code extension and place the .vsix file
build_vscode_extension()
# Poetry will handle including files based on pyproject.toml `include` patterns.
# Ensure openhands/integrations/vscode/*.vsix is included there.
print('--- Custom Poetry build script (build_vscode.py) finished ---')
if __name__ == '__main__':
print('Running build_vscode.py directly for testing VS Code extension packaging...')
build_vscode_extension()
print('Direct execution of build_vscode.py finished.')

View File

@@ -440,12 +440,6 @@ type = "noop"
#temperature = 0.1
#max_input_tokens = 1024
#################################### Eval ####################################
# Configuration for the evaluation, please refer to the specific evaluation
# plugin for the available options
##############################################################################
########################### Kubernetes #######################################
# Kubernetes configuration when using the Kubernetes runtime
##############################################################################

View File

@@ -1,5 +1,5 @@
ARG OPENHANDS_BUILD_VERSION=dev
FROM node:24.8-trixie-slim AS frontend-builder
FROM node:25.2-trixie-slim AS frontend-builder
WORKDIR /app

View File

@@ -12,7 +12,8 @@ services:
- SANDBOX_API_HOSTNAME=host.docker.internal
- DOCKER_HOST_ADDR=host.docker.internal
#
- SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-ghcr.io/openhands/runtime:1.0-nikolaik}
- AGENT_SERVER_IMAGE_REPOSITORY=${AGENT_SERVER_IMAGE_REPOSITORY:-ghcr.io/openhands/agent-server}
- AGENT_SERVER_IMAGE_TAG=${AGENT_SERVER_IMAGE_TAG:-1.12.0-python}
- SANDBOX_USER_ID=${SANDBOX_USER_ID:-1234}
- WORKSPACE_MOUNT_PATH=${WORKSPACE_BASE:-$PWD/workspace}
ports:

View File

@@ -10,6 +10,15 @@ repos:
args: ["--allow-multiple-documents"]
- id: debug-statements
- repo: local
hooks:
- id: warn-appmode-oss
name: "Warn on AppMode.OSS in backend (use AppMode.OPENHANDS)"
language: system
entry: bash -lc 'if rg -n "\\bAppMode\\.OSS\\b" openhands tests/unit; then echo "Found AppMode.OSS usage. Prefer AppMode.OPENHANDS."; exit 1; fi'
pass_filenames: false
- repo: https://github.com/tox-dev/pyproject-fmt
rev: v2.5.1
hooks:

View File

@@ -7,7 +7,8 @@ services:
image: openhands:latest
container_name: openhands-app-${DATE:-}
environment:
- SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-docker.openhands.dev/openhands/runtime:1.0-nikolaik}
- AGENT_SERVER_IMAGE_REPOSITORY=${AGENT_SERVER_IMAGE_REPOSITORY:-ghcr.io/openhands/agent-server}
- AGENT_SERVER_IMAGE_TAG=${AGENT_SERVER_IMAGE_TAG:-1.12.0-python}
#- SANDBOX_USER_ID=${SANDBOX_USER_ID:-1234} # enable this only if you want a specific non-root sandbox user but you will have to manually adjust permissions of ~/.openhands for this user
- WORKSPACE_MOUNT_PATH=${WORKSPACE_BASE:-$PWD/workspace}
ports:

View File

@@ -23,12 +23,23 @@ RUN apt-get update && \
apt-get clean && \
rm -rf /var/lib/apt/lists/*
# Install Python packages with security fixes
RUN pip install alembic psycopg2-binary cloud-sql-python-connector pg8000 gspread stripe python-keycloak asyncpg sqlalchemy[asyncio] resend tenacity slack-sdk ddtrace "posthog>=6.0.0" "limits==5.2.0" coredis prometheus-client shap scikit-learn pandas numpy && \
# Update packages with known CVE fixes
pip install --upgrade \
"mcp>=1.10.0" \
"pillow>=11.3.0"
# Install poetry and export before importing current code.
RUN /app/.venv/bin/pip install poetry poetry-plugin-export
# Install Python dependencies from poetry.lock for reproducible builds
# Copy lock files first for better Docker layer caching
COPY --chown=openhands:openhands enterprise/pyproject.toml enterprise/poetry.lock /tmp/enterprise/
RUN cd /tmp/enterprise && \
# Export only main dependencies with hashes for supply chain security
/app/.venv/bin/poetry export --only main -o requirements.txt && \
# Remove the local path dependency (openhands-ai is already in base image)
sed -i '/^-e /d; /openhands-ai/d' requirements.txt && \
# Install pinned dependencies from lock file
/app/.venv/bin/pip install -r requirements.txt && \
# Cleanup - return to /app before removing /tmp/enterprise
cd /app && \
rm -rf /tmp/enterprise && \
/app/.venv/bin/pip uninstall -y poetry poetry-plugin-export
WORKDIR /app
COPY --chown=openhands:openhands --chmod=770 enterprise .

View File

@@ -2,7 +2,7 @@ BACKEND_HOST ?= "127.0.0.1"
BACKEND_PORT = 3000
BACKEND_HOST_PORT = "$(BACKEND_HOST):$(BACKEND_PORT)"
FRONTEND_PORT = 3001
OPENHANDS_PATH ?= "../../OpenHands"
OPENHANDS_PATH ?= ".."
OPENHANDS := $(OPENHANDS_PATH)
OPENHANDS_FRONTEND_PATH = $(OPENHANDS)/frontend/build

View File

@@ -1,6 +1,6 @@
# OpenHands Enterprise Server
> [!WARNING]
> This software is licensed under the [Polyform Free Trial License](./LICENSE). This is **NOT** an open source license. Usage is limited to 30 days per calendar year without a commercial license. If you would like to use it beyond 30 days, please [contact us](https://www.all-hands.dev/contact).
> This software is licensed under the [Polyform Free Trial License](./LICENSE). This is **NOT** an open source license. Usage is limited to 30 days per calendar year without a commercial license. If you would like to use it beyond 30 days, please [contact us](https://www.openhands.dev/contact).
> [!WARNING]
> This is a work in progress and may contain bugs, incomplete features, or breaking changes.
@@ -10,13 +10,13 @@ This directory contains the enterprise server used by [OpenHands Cloud](https://
You may also want to check out the MIT-licensed [OpenHands](https://github.com/OpenHands/OpenHands)
## Extension of OpenHands (OSS)
## Extension of OpenHands
The code in `/enterprise` directory builds on top of open source (OSS) code, extending its functionality. The enterprise code is entangled with the OSS code in two ways
The code in `/enterprise` builds on top of OpenHands (MIT-licensed), extending its functionality. The enterprise code is entangled with OpenHands in two ways:
- Enterprise stacks on top of OSS. For example, the middleware in enterprise is stacked right on top of the middlewares in OSS. In `SAAS`, the middleware from BOTH repos will be present and running (which can sometimes cause conflicts)
- Enterprise stacks on top of OpenHands. For example, the middleware in enterprise is stacked right on top of the middlewares in OpenHands. In `SAAS`, the middleware from BOTH repos will be present and running (which can sometimes cause conflicts)
- Enterprise overrides the implementation in OSS (only one is present at a time). For example, the server config SaasServerConfig which overrides [`ServerConfig`](https://github.com/OpenHands/OpenHands/blob/main/openhands/server/config/server_config.py#L8) on OSS. This is done through dynamic imports ([see here](https://github.com/OpenHands/OpenHands/blob/main/openhands/server/config/server_config.py#L37-#L45))
- Enterprise overrides the implementation in OpenHands (only one is present at a time). For example, the server config SaasServerConfig overrides [`ServerConfig`](https://github.com/OpenHands/OpenHands/blob/main/openhands/server/config/server_config.py#L8) in OpenHands. This is done through dynamic imports ([see here](https://github.com/OpenHands/OpenHands/blob/main/openhands/server/config/server_config.py#L37-#L45))
Key areas that change on `SAAS` are
@@ -26,11 +26,11 @@ Key areas that change on `SAAS` are
### Authentication
| Aspect | OSS | Enterprise |
| Aspect | OpenHands | Enterprise |
| ------------------------- | ------------------------------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------- |
| **Authentication Method** | User adds a personal access token (PAT) through the UI | User performs OAuth through the UI. The Github app provides a short-lived access token and refresh token |
| **Authentication Method** | User adds a personal access token (PAT) through the UI | User performs OAuth through the UI. The GitHub app provides a short-lived access token and refresh token |
| **Token Storage** | PAT is stored in **Settings** | Token is stored in **GithubTokenManager** (a file store in our backend) |
| **Authenticated status** | We simply check if token exists in `Settings` | We issue a signed cookie with `github_user_id` during oauth, so subsequent requests with the cookie can be considered authenticated |
| **Authenticated status** | We simply check if token exists in `Settings` | We issue a signed cookie with `github_user_id` during OAuth, so subsequent requests with the cookie can be considered authenticated |
Note that in the future, authentication will happen via keycloak. All modifications for authentication will happen in enterprise.
@@ -38,7 +38,7 @@ Note that in the future, authentication will happen via keycloak. All modificati
The github service is responsible for interacting with Github APIs. As a consequence, it uses the user's token and refreshes it if need be
| Aspect | OSS | Enterprise |
| Aspect | OpenHands | Enterprise |
| ------------------------- | -------------------------------------- | ---------------------------------------------- |
| **Class used** | `GitHubService` | `SaaSGitHubService` |
| **Token used** | User's PAT fetched from `Settings` | User's token fetched from `GitHubTokenManager` |
@@ -50,7 +50,7 @@ NOTE: in the future we will simply replace the `GithubTokenManager` with keycloa
## User ID vs User Token
- On OSS, the entire APP revolves around the Github token the user sets. `openhands/server` uses `request.state.github_token` for the entire app
- On Enterprise, the entire APP resolves around the Github User ID. This is because the cookie sets it, so `openhands/server` AND `enterprise/server` depend on it and completly ignore `request.state.github_token` (token is fetched from `GithubTokenManager` instead)
- In OpenHands, the entire app revolves around the GitHub token the user sets. `openhands/server` uses `request.state.github_token` for the entire app
- On Enterprise, the entire APP resolves around the Github User ID. This is because the cookie sets it, so `openhands/server` AND `enterprise/server` depend on it and completely ignore `request.state.github_token` (token is fetched from `GithubTokenManager` instead)
Note that introducing Github User ID on OSS, for instance, will cause large breakages.
Note that introducing GitHub User ID in OpenHands, for instance, will cause large breakages.

View File

@@ -1772,6 +1772,40 @@
"sendIdTokenOnLogout": "true",
"passMaxAge": "false"
}
},
{
"alias": "bitbucket_data_center",
"displayName": "Bitbucket Data Center",
"internalId": "b77b4ead-20e8-451c-ad27-99f92d561616",
"providerId": "oauth2",
"enabled": true,
"updateProfileFirstLoginMode": "on",
"trustEmail": true,
"storeToken": true,
"addReadTokenRoleOnCreate": false,
"authenticateByDefault": false,
"linkOnly": false,
"hideOnLogin": false,
"config": {
"givenNameClaim": "given_name",
"userInfoUrl": "https://${WEB_HOST}/bitbucket-dc-proxy/oauth2/userinfo",
"clientId": "$BITBUCKET_DATA_CENTER_CLIENT_ID",
"tokenUrl": "https://${BITBUCKET_DATA_CENTER_HOST}/rest/oauth2/latest/token",
"acceptsPromptNoneForwardFromClient": "false",
"fullNameClaim": "name",
"userIDClaim": "sub",
"emailClaim": "email",
"userNameClaim": "preferred_username",
"caseSensitiveOriginalUsername": "false",
"familyNameClaim": "family_name",
"pkceEnabled": "false",
"authorizationUrl": "https://${BITBUCKET_DATA_CENTER_HOST}/rest/oauth2/latest/authorize",
"clientAuthMethod": "client_secret_post",
"syncMode": "IMPORT",
"clientSecret": "$BITBUCKET_DATA_CENTER_CLIENT_SECRET",
"allowedClockSkew": "0",
"defaultScope": "REPO_WRITE"
}
}
],
"identityProviderMappers": [
@@ -1829,6 +1863,26 @@
"syncMode": "FORCE",
"attribute": "identity_provider"
}
},
{
"name": "id-mapper",
"identityProviderAlias": "bitbucket_data_center",
"identityProviderMapper": "oidc-user-attribute-idp-mapper",
"config": {
"syncMode": "FORCE",
"claim": "sub",
"user.attribute": "bitbucket_data_center_id"
}
},
{
"name": "identity-provider",
"identityProviderAlias": "bitbucket_data_center",
"identityProviderMapper": "hardcoded-attribute-idp-mapper",
"config": {
"attribute.value": "bitbucket_data_center",
"syncMode": "FORCE",
"attribute": "identity_provider"
}
}
],
"components": {

View File

@@ -50,8 +50,10 @@ repos:
- ./
- stripe==11.5.0
- pygithub==2.6.1
# To see gaps add `--html-report mypy-report/`
entry: mypy --config-file enterprise/dev_config/python/mypy.ini enterprise/
# Use -p (package) to avoid dual module name conflict when using MYPYPATH
# MYPYPATH=enterprise allows resolving bare imports like "from integrations.xxx"
# Note: tests package excluded to avoid conflict with core openhands tests
entry: bash -c 'MYPYPATH=enterprise mypy --config-file enterprise/dev_config/python/mypy.ini -p integrations -p server -p storage -p sync'
always_run: true
pass_filenames: false
files: ^enterprise/

View File

@@ -2,7 +2,6 @@
warn_unused_configs = True
ignore_missing_imports = True
check_untyped_defs = True
explicit_package_bases = True
warn_unreachable = True
warn_redundant_casts = True
no_implicit_optional = True

View File

@@ -0,0 +1,13 @@
# Enterprise Architecture Documentation
Architecture diagrams specific to the OpenHands SaaS/Enterprise deployment.
## Documentation
- [Authentication Flow](./authentication.md) - Keycloak-based authentication for SaaS deployment
- [External Integrations](./external-integrations.md) - GitHub, Slack, Jira, and other service integrations
## Related Documentation
For core OpenHands architecture (applicable to all deployments), see:
- [Core Architecture Documentation](../../../openhands/architecture/README.md)

View File

@@ -0,0 +1,58 @@
# Authentication Flow (SaaS Deployment)
OpenHands uses Keycloak for identity management in the SaaS deployment. The authentication flow involves multiple services:
```mermaid
sequenceDiagram
autonumber
participant User as User (Browser)
participant App as App Server
participant KC as Keycloak
participant IdP as Identity Provider<br/>(GitHub, Google, etc.)
participant DB as User Database
Note over User,DB: OAuth 2.0 / OIDC Authentication Flow
User->>App: Access OpenHands
App->>User: Redirect to Keycloak
User->>KC: Login request
KC->>User: Show login options
User->>KC: Select provider (e.g., GitHub)
KC->>IdP: OAuth redirect
User->>IdP: Authenticate
IdP-->>KC: OAuth callback + tokens
Note over KC: Create/update user session
KC-->>User: Redirect with auth code
User->>App: Auth code
App->>KC: Exchange code for tokens
KC-->>App: Access token + Refresh token
Note over App: Create signed JWT cookie
App->>DB: Store/update user record
App-->>User: Set keycloak_auth cookie
Note over User,DB: Subsequent Requests
User->>App: Request with cookie
Note over App: Verify JWT signature
App->>KC: Validate token (if needed)
KC-->>App: Token valid
Note over App: Extract user context
App-->>User: Authorized response
```
### Authentication Components
| Component | Purpose | Location |
|-----------|---------|----------|
| **Keycloak** | Identity provider, SSO, token management | External service |
| **UserAuth** | Abstract auth interface | `openhands/server/user_auth/user_auth.py` |
| **SaasUserAuth** | Keycloak implementation | `enterprise/server/auth/saas_user_auth.py` |
| **JWT Service** | Token signing/verification | `openhands/app_server/services/jwt_service.py` |
| **Auth Routes** | Login/logout endpoints | `enterprise/server/routes/auth.py` |
### Token Flow
1. **Keycloak Access Token**: Short-lived token for API access
2. **Keycloak Refresh Token**: Long-lived token to obtain new access tokens
3. **Signed JWT Cookie**: App Server's session cookie containing encrypted Keycloak tokens
4. **Provider Tokens**: OAuth tokens for GitHub, GitLab, etc. (stored separately for git operations)

View File

@@ -0,0 +1,88 @@
# External Integrations
OpenHands integrates with external services (GitHub, Slack, Jira, etc.) through webhook-based event handling:
```mermaid
sequenceDiagram
autonumber
participant Ext as External Service<br/>(GitHub/Slack/Jira)
participant App as App Server
participant IntRouter as Integration Router
participant Manager as Integration Manager
participant Conv as Conversation Service
participant Sandbox as Sandbox
Note over Ext,Sandbox: Webhook Event Flow (e.g., GitHub Issue Created)
Ext->>App: POST /api/integration/{service}/events
App->>IntRouter: Route to service handler
Note over IntRouter: Verify signature (HMAC)
IntRouter->>Manager: Parse event payload
Note over Manager: Extract context (repo, issue, user)
Note over Manager: Map external user → OpenHands user
Manager->>Conv: Create conversation (with issue context)
Conv->>Sandbox: Provision sandbox
Sandbox-->>Conv: Ready
Manager->>Sandbox: Start agent with task
Note over Ext,Sandbox: Agent Works on Task...
Sandbox-->>Manager: Task complete
Manager->>Ext: POST result<br/>(PR, comment, etc.)
Note over Ext,Sandbox: Callback Flow (Agent → External Service)
Sandbox->>App: Webhook callback<br/>/api/v1/webhooks
App->>Manager: Process callback
Manager->>Ext: Update external service
```
### Supported Integrations
| Integration | Trigger Events | Agent Actions |
|-------------|----------------|---------------|
| **GitHub** | Issue created, PR opened, @mention | Create PR, comment, push commits |
| **GitLab** | Issue created, MR opened | Create MR, comment, push commits |
| **Slack** | @mention in channel | Reply in thread, create tasks |
| **Jira** | Issue created/updated | Update ticket, add comments |
| **Linear** | Issue created | Update status, add comments |
### Integration Components
| Component | Purpose | Location |
|-----------|---------|----------|
| **Integration Routes** | Webhook endpoints per service | `enterprise/server/routes/integration/` |
| **Integration Managers** | Business logic per service | `enterprise/integrations/{service}/` |
| **Token Manager** | Store/retrieve OAuth tokens | `enterprise/server/auth/token_manager.py` |
| **Callback Processor** | Handle agent → service updates | `enterprise/integrations/{service}/*_callback_processor.py` |
### Integration Authentication
```
External Service (e.g., GitHub)
┌─────────────────────────────────┐
│ GitHub App Installation │
│ - Webhook secret for signature │
│ - App private key for API calls │
└─────────────────────────────────┘
┌─────────────────────────────────┐
│ User Account Linking │
│ - Keycloak user ID │
│ - GitHub user ID │
│ - Stored OAuth tokens │
└─────────────────────────────────┘
┌─────────────────────────────────┐
│ Agent Execution │
│ - Uses linked tokens for API │
│ - Can push, create PRs, comment │
└─────────────────────────────────┘
```

View File

@@ -200,7 +200,7 @@ class MetricsCollector(ABC):
"""Base class for metrics collectors."""
@abstractmethod
def collect(self) -> List[MetricResult]:
async def collect(self) -> List[MetricResult]:
"""Collect metrics and return results."""
pass
@@ -264,12 +264,13 @@ class SystemMetricsCollector(MetricsCollector):
def collector_name(self) -> str:
return "system_metrics"
def collect(self) -> List[MetricResult]:
async def collect(self) -> List[MetricResult]:
results = []
# Collect user count
with session_maker() as session:
user_count = session.query(UserSettings).count()
async with a_session_maker() as session:
user_count_result = await session.execute(select(func.count()).select_from(UserSettings))
user_count = user_count_result.scalar()
results.append(MetricResult(
key="total_users",
value=user_count
@@ -277,9 +278,11 @@ class SystemMetricsCollector(MetricsCollector):
# Collect conversation count (last 30 days)
thirty_days_ago = datetime.now(timezone.utc) - timedelta(days=30)
conversation_count = session.query(StoredConversationMetadata)\
.filter(StoredConversationMetadata.created_at >= thirty_days_ago)\
.count()
conversation_count_result = await session.execute(
select(func.count()).select_from(StoredConversationMetadata)
.where(StoredConversationMetadata.created_at >= thirty_days_ago)
)
conversation_count = conversation_count_result.scalar()
results.append(MetricResult(
key="conversations_30d",
@@ -303,7 +306,7 @@ class TelemetryCollectionProcessor(MaintenanceTaskProcessor):
"""Collect metrics from all registered collectors."""
# Check if collection is needed
if not self._should_collect():
if not await self._should_collect():
return {"status": "skipped", "reason": "too_recent"}
# Collect metrics from all registered collectors
@@ -313,7 +316,7 @@ class TelemetryCollectionProcessor(MaintenanceTaskProcessor):
for collector in collector_registry.get_all_collectors():
try:
if collector.should_collect():
results = collector.collect()
results = await collector.collect()
for result in results:
all_metrics[result.key] = result.value
collector_results[collector.collector_name] = len(results)
@@ -322,13 +325,13 @@ class TelemetryCollectionProcessor(MaintenanceTaskProcessor):
collector_results[collector.collector_name] = f"error: {e}"
# Store metrics in database
with session_maker() as session:
async with a_session_maker() as session:
telemetry_record = TelemetryMetrics(
metrics_data=all_metrics,
collected_at=datetime.now(timezone.utc)
)
session.add(telemetry_record)
session.commit()
await session.commit()
# Note: No need to track last_collection_at separately
# Can be derived from MAX(collected_at) in telemetry_metrics
@@ -339,11 +342,12 @@ class TelemetryCollectionProcessor(MaintenanceTaskProcessor):
"collectors_run": collector_results
}
def _should_collect(self) -> bool:
async def _should_collect(self) -> bool:
"""Check if collection is needed based on interval."""
with session_maker() as session:
async with a_session_maker() as session:
# Get last collection time from metrics table
last_collected = session.query(func.max(TelemetryMetrics.collected_at)).scalar()
result = await session.execute(select(func.max(TelemetryMetrics.collected_at)))
last_collected = result.scalar()
if not last_collected:
return True
@@ -366,17 +370,19 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
"""Upload pending metrics to Replicated."""
# Get pending metrics
with session_maker() as session:
pending_metrics = session.query(TelemetryMetrics)\
.filter(TelemetryMetrics.uploaded_at.is_(None))\
.order_by(TelemetryMetrics.collected_at)\
.all()
async with a_session_maker() as session:
result = await session.execute(
select(TelemetryMetrics)
.where(TelemetryMetrics.uploaded_at.is_(None))
.order_by(TelemetryMetrics.collected_at)
)
pending_metrics = result.scalars().all()
if not pending_metrics:
return {"status": "no_pending_metrics"}
# Get admin email - skip if not available
admin_email = self._get_admin_email()
admin_email = await self._get_admin_email()
if not admin_email:
logger.info("Skipping telemetry upload - no admin email available")
return {
@@ -413,13 +419,15 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
await instance.set_status(InstanceStatus.RUNNING)
# Mark as uploaded
with session_maker() as session:
record = session.query(TelemetryMetrics)\
.filter(TelemetryMetrics.id == metric_record.id)\
.first()
async with a_session_maker() as session:
result = await session.execute(
select(TelemetryMetrics)
.where(TelemetryMetrics.id == metric_record.id)
)
record = result.scalar_one_or_none()
if record:
record.uploaded_at = datetime.now(timezone.utc)
session.commit()
await session.commit()
uploaded_count += 1
@@ -427,14 +435,16 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
logger.error(f"Failed to upload metrics {metric_record.id}: {e}")
# Update error info
with session_maker() as session:
record = session.query(TelemetryMetrics)\
.filter(TelemetryMetrics.id == metric_record.id)\
.first()
async with a_session_maker() as session:
result = await session.execute(
select(TelemetryMetrics)
.where(TelemetryMetrics.id == metric_record.id)
)
record = result.scalar_one_or_none()
if record:
record.upload_attempts += 1
record.last_upload_error = str(e)
session.commit()
await session.commit()
failed_count += 1
@@ -448,7 +458,7 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
"total_processed": len(pending_metrics)
}
def _get_admin_email(self) -> str | None:
async def _get_admin_email(self) -> str | None:
"""Get administrator email for customer identification."""
# 1. Check environment variable first
env_admin_email = os.getenv('OPENHANDS_ADMIN_EMAIL')
@@ -457,12 +467,15 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
return env_admin_email
# 2. Use first active user's email (earliest accepted_tos)
with session_maker() as session:
first_user = session.query(UserSettings)\
.filter(UserSettings.email.isnot(None))\
.filter(UserSettings.accepted_tos.isnot(None))\
.order_by(UserSettings.accepted_tos.asc())\
.first()
async with a_session_maker() as session:
result = await session.execute(
select(UserSettings)
.where(UserSettings.email.isnot(None))
.where(UserSettings.accepted_tos.isnot(None))
.order_by(UserSettings.accepted_tos.asc())
.limit(1)
)
first_user = result.scalar_one_or_none()
if first_user and first_user.email:
logger.info(f"Using first active user email: {first_user.email}")
@@ -474,15 +487,16 @@ class TelemetryUploadProcessor(MaintenanceTaskProcessor):
async def _update_telemetry_identity(self, customer_id: str, instance_id: str) -> None:
"""Update or create telemetry identity record."""
with session_maker() as session:
identity = session.query(TelemetryIdentity).first()
async with a_session_maker() as session:
result = await session.execute(select(TelemetryIdentity).limit(1))
identity = result.scalar_one_or_none()
if not identity:
identity = TelemetryIdentity()
session.add(identity)
identity.customer_id = customer_id
identity.instance_id = instance_id
session.commit()
await session.commit()
```
### 4.4 License Warning System
@@ -503,11 +517,13 @@ async def get_license_status():
if not _is_openhands_enterprise():
return {"warn": False, "message": ""}
with session_maker() as session:
async with a_session_maker() as session:
# Get last successful upload time from metrics table
last_upload = session.query(func.max(TelemetryMetrics.uploaded_at))\
.filter(TelemetryMetrics.uploaded_at.isnot(None))\
.scalar()
result = await session.execute(
select(func.max(TelemetryMetrics.uploaded_at))
.where(TelemetryMetrics.uploaded_at.isnot(None))
)
last_upload = result.scalar()
if not last_upload:
# No successful uploads yet - show warning after 4 days
@@ -521,10 +537,13 @@ async def get_license_status():
if days_since_upload > 4:
# Find oldest unsent batch
oldest_unsent = session.query(TelemetryMetrics)\
.filter(TelemetryMetrics.uploaded_at.is_(None))\
.order_by(TelemetryMetrics.collected_at)\
.first()
result = await session.execute(
select(TelemetryMetrics)
.where(TelemetryMetrics.uploaded_at.is_(None))
.order_by(TelemetryMetrics.collected_at)
.limit(1)
)
oldest_unsent = result.scalar_one_or_none()
if oldest_unsent:
# Calculate expiration date (oldest unsent + 34 days)
@@ -630,19 +649,23 @@ spec:
- python
- -c
- |
import asyncio
from enterprise.storage.maintenance_task import MaintenanceTask, MaintenanceTaskStatus
from enterprise.storage.database import session_maker
from enterprise.storage.database import a_session_maker
from enterprise.server.telemetry.collection_processor import TelemetryCollectionProcessor
# Create collection task
processor = TelemetryCollectionProcessor()
task = MaintenanceTask()
task.set_processor(processor)
task.status = MaintenanceTaskStatus.PENDING
async def main():
# Create collection task
processor = TelemetryCollectionProcessor()
task = MaintenanceTask()
task.set_processor(processor)
task.status = MaintenanceTaskStatus.PENDING
with session_maker() as session:
session.add(task)
session.commit()
async with a_session_maker() as session:
session.add(task)
await session.commit()
asyncio.run(main())
restartPolicy: OnFailure
```
@@ -680,23 +703,27 @@ spec:
- python
- -c
- |
import asyncio
from enterprise.storage.maintenance_task import MaintenanceTask, MaintenanceTaskStatus
from enterprise.storage.database import session_maker
from enterprise.storage.database import a_session_maker
from enterprise.server.telemetry.upload_processor import TelemetryUploadProcessor
import os
# Create upload task
processor = TelemetryUploadProcessor(
replicated_publishable_key=os.getenv('REPLICATED_PUBLISHABLE_KEY'),
replicated_app_slug=os.getenv('REPLICATED_APP_SLUG', 'openhands-enterprise')
)
task = MaintenanceTask()
task.set_processor(processor)
task.status = MaintenanceTaskStatus.PENDING
async def main():
# Create upload task
processor = TelemetryUploadProcessor(
replicated_publishable_key=os.getenv('REPLICATED_PUBLISHABLE_KEY'),
replicated_app_slug=os.getenv('REPLICATED_APP_SLUG', 'openhands-enterprise')
)
task = MaintenanceTask()
task.set_processor(processor)
task.status = MaintenanceTaskStatus.PENDING
with session_maker() as session:
session.add(task)
session.commit()
async with a_session_maker() as session:
session.add(task)
await session.commit()
asyncio.run(main())
restartPolicy: OnFailure
```

View File

@@ -0,0 +1,131 @@
# Plugin Launch Flow
This document describes how plugins are launched in OpenHands Saas / Enterprise, from the plugin directory through to agent execution.
## Architecture Overview
```
Plugin Directory ──▶ Frontend /launch ──▶ App Server ──▶ Agent Server ──▶ SDK
(external) (modal) (API) (in sandbox) (plugin loading)
```
| Component | Responsibility |
|-----------|---------------|
| **Plugin Directory** | Index plugins, present to user, construct launch URLs |
| **Frontend** | Display confirmation modal, collect parameters, call API |
| **App Server** | Validate request, pass plugin specs to agent server |
| **Agent Server** | Run inside sandbox, delegate plugin loading to SDK |
| **SDK** | Fetch plugins, load contents, merge skills/hooks/MCP into agent |
## User Experience
### Plugin Directory
The plugin directory presents users with a catalog of available plugins. For each plugin, users see:
- Plugin name and description (from `plugin.json`)
- Author and version information
- A "Launch" button
When a user clicks "Launch", the plugin directory:
1. Reads the plugin's `entry_command` to know which slash command to invoke
2. Determines what parameters the plugin accepts (if any)
3. Redirects to OpenHands with this information encoded in the URL
### Parameter Collection
If a plugin requires user input (API keys, configuration values, etc.), the frontend displays a form modal before starting the conversation. Parameters are passed in the launch URL and rendered as form fields based on their type:
- **String values** → Text input
- **Number values** → Number input
- **Boolean values** → Checkbox
Only primitive types are supported. Complex types (arrays, objects) are not currently supported for parameter input.
The user fills in required values, then clicks "Start Conversation" to proceed.
## Launch Flow
1. **Plugin Directory** (external) constructs a launch URL to the OpenHands app server when user clicks "Launch":
```
/launch?plugins=BASE64_JSON&message=/city-weather:now%20Tokyo
```
The `plugins` parameter includes any parameter definitions with default values:
```json
[{
"source": "github:owner/repo",
"repo_path": "plugins/my-plugin",
"parameters": {"api_key": "", "timeout": 30, "debug": false}
}]
```
2. **OpenHands Frontend** (`/launch` route, [PR #12699](https://github.com/OpenHands/OpenHands/pull/12699)) displays modal with parameter form, collects user input
3. **OpenHands App Server** ([PR #12338](https://github.com/OpenHands/OpenHands/pull/12338)) receives the API call:
```
POST /api/v1/app-conversations
{
"plugins": [{"source": "github:owner/repo", "repo_path": "plugins/city-weather"}],
"initial_message": {"content": [{"type": "text", "text": "/city-weather:now Tokyo"}]}
}
```
Call stack:
- `AppConversationRouter` receives request with `PluginSpec` list
- `LiveStatusAppConversationService._finalize_conversation_request()` converts `PluginSpec` → `PluginSource`
- Creates `StartConversationRequest(plugins=sdk_plugins, ...)` and sends to agent server
4. **Agent Server** (inside sandbox, [SDK PR #1651](https://github.com/OpenHands/software-agent-sdk/pull/1651)) stores specs, defers loading:
Call stack:
- `ConversationService.start_conversation()` receives `StartConversationRequest`
- Creates `StoredConversation` with plugin specs
- Creates `LocalConversation(plugins=request.plugins, ...)`
- Plugin loading deferred until first `run()` or `send_message()`
5. **SDK** fetches and loads plugins on first use:
Call stack:
- `LocalConversation._ensure_plugins_loaded()` triggered by first message
- For each plugin spec:
- `Plugin.fetch(source, ref, repo_path)` → clones/caches git repo
- `Plugin.load(path)` → parses `plugin.json`, loads commands/skills/hooks
- `plugin.add_skills_to(context)` → merges skills into agent
- `plugin.add_mcp_config_to(config)` → merges MCP servers
6. **Agent** receives message, `/city-weather:now` triggers the skill
## Key Design Decisions
### Plugin Loading in Sandbox
Plugins load **inside the sandbox** because:
- Plugin hooks and scripts need isolated execution
- MCP servers run inside the sandbox
- Skills may reference sandbox filesystem
### Entry Command Handling
The `entry_command` field in `plugin.json` allows plugin authors to declare a default command:
```json
{
"name": "city-weather",
"entry_command": "now"
}
```
This flows through the system:
1. Plugin author declares `entry_command` in plugin.json
2. Plugin directory reads it when indexing
3. Plugin directory includes `/city-weather:now` in the launch URL's `message` parameter
4. Message passes through to agent as `initial_message`
The SDK exposes this field but does not auto-invoke it—callers control the initial message.
## Related
- [OpenHands PR #12338](https://github.com/OpenHands/OpenHands/pull/12338) - App server plugin support
- [OpenHands PR #12699](https://github.com/OpenHands/OpenHands/pull/12699) - Frontend `/launch` route
- [SDK PR #1651](https://github.com/OpenHands/software-agent-sdk/pull/1651) - Agent server plugin loading
- [SDK PR #1647](https://github.com/OpenHands/software-agent-sdk/pull/1647) - Plugin.fetch() for remote plugin fetching

View File

@@ -2,7 +2,7 @@
You have a few options here, which are expanded on below:
- A simple local development setup, with live reloading for both OSS and this repo
- A simple local development setup, with live reloading for both OpenHands and this repo
- A more complex setup that includes Redis
- An even more complex setup that includes GitHub events
@@ -26,7 +26,7 @@ Before starting, make sure you have the following tools installed:
## Option 1: Simple local development
This option will allow you to modify the both the OSS code and the code in this repo,
This option will allow you to modify both the OpenHands code and the code in this repo,
and see the changes in real-time.
This option works best for most scenarios. The only thing it's missing is
@@ -50,7 +50,7 @@ First run this to retrieve Github App secrets
```
gcloud auth application-default login
gcloud config set project global-432717
local/decrypt_env.sh
enterprise_local/decrypt_env.sh /path/to/root/of/deploy/repo
```
Now run this to generate a `.env` file, which will used to run SAAS locally
@@ -105,9 +105,9 @@ export REDIS_PORT=6379
(see above)
### 2. Build OSS Openhands
### 2. Build OpenHands
Develop on [Openhands](https://github.com/All-Hands-AI/OpenHands) locally. When ready, run the following inside Openhands repo (not the Deploy repo)
Develop on [Openhands](https://github.com/OpenHands/OpenHands) locally. When ready, run the following inside Openhands repo (not the Deploy repo)
```
docker build -f containers/app/Dockerfile -t openhands .
@@ -155,7 +155,7 @@ Visit the tunnel domain found in Step 4 to run the app (`https://bc71-2603-7000-
### Local Debugging with VSCode
Local Development necessitates running a version of OpenHands that is as similar as possible to the version running in the SAAS Environment. Before running these steps, it is assumed you have a local development version of the OSS OpenHands project running.
Local Development necessitates running a version of OpenHands that is as similar as possible to the version running in the SAAS Environment. Before running these steps, it is assumed you have a local development version of OpenHands running.
#### Redis
@@ -201,8 +201,8 @@ And then invoking `printenv`. NOTE: _DO NOT DO THIS WITH PROD!!!_ (Hopefully by
"DEBUG": "1",
"FILE_STORE": "local",
"REDIS_HOST": "localhost:6379",
"OPENHANDS": "<YOUR LOCAL OSS OPENHANDS DIR>",
"FRONTEND_DIRECTORY": "<YOUR LOCAL OSS OPENHANDS DIR>/frontend/build",
"OPENHANDS": "<YOUR LOCAL OPENHANDS DIR>",
"FRONTEND_DIRECTORY": "<YOUR LOCAL OPENHANDS DIR>/frontend/build",
"SANDBOX_RUNTIME_CONTAINER_IMAGE": "ghcr.io/openhands/runtime:main-nikolaik",
"FILE_STORE_PATH": "<YOUR HOME DIRECTORY>>/.openhands-state",
"OPENHANDS_CONFIG_CLS": "server.config.SaaSServerConfig",
@@ -235,8 +235,8 @@ And then invoking `printenv`. NOTE: _DO NOT DO THIS WITH PROD!!!_ (Hopefully by
"DEBUG": "1",
"FILE_STORE": "local",
"REDIS_HOST": "localhost:6379",
"OPENHANDS": "<YOUR LOCAL OSS OPENHANDS DIR>",
"FRONTEND_DIRECTORY": "<YOUR LOCAL OSS OPENHANDS DIR>/frontend/build",
"OPENHANDS": "<YOUR LOCAL OPENHANDS DIR>",
"FRONTEND_DIRECTORY": "<YOUR LOCAL OPENHANDS DIR>/frontend/build",
"SANDBOX_RUNTIME_CONTAINER_IMAGE": "ghcr.io/openhands/runtime:main-nikolaik",
"FILE_STORE_PATH": "<YOUR HOME DIRECTORY>>/.openhands-state",
"OPENHANDS_CONFIG_CLS": "server.config.SaaSServerConfig",

View File

@@ -109,6 +109,9 @@ lines.append(
lines.append(
'OPENHANDS_BITBUCKET_SERVICE_CLS=integrations.bitbucket.bitbucket_service.SaaSBitBucketService'
)
lines.append(
'OPENHANDS_BITBUCKET_DATA_CENTER_SERVICE_CLS=integrations.bitbucket_data_center.bitbucket_dc_service.SaaSBitbucketDCService'
)
lines.append(
'OPENHANDS_CONVERSATION_VALIDATOR_CLS=storage.saas_conversation_validator.SaasConversationValidator'
)

4
enterprise/enterprise_local/decrypt_env.sh Normal file → Executable file
View File

@@ -4,12 +4,12 @@ set -euo pipefail
# Check if DEPLOY_DIR argument was provided
if [ $# -lt 1 ]; then
echo "Usage: $0 <DEPLOY_DIR>"
echo "Example: $0 /path/to/deploy"
echo "Example: $0 /path/to/root/of/deploy/repo"
exit 1
fi
# Normalize path (remove trailing slash)
DEPLOY_DIR="${DEPLOY_DIR%/}"
DEPLOY_DIR="${1%/}"
# Function to decrypt and rename
decrypt_and_move() {

View File

@@ -1,47 +0,0 @@
import os
import posthog
from openhands.core.logger import openhands_logger as logger
# Initialize PostHog
posthog.api_key = os.environ.get('POSTHOG_CLIENT_KEY', 'phc_placeholder')
posthog.host = os.environ.get('POSTHOG_HOST', 'https://us.i.posthog.com')
# Log PostHog configuration with masked API key for security
api_key = posthog.api_key
if api_key and len(api_key) > 8:
masked_key = f'{api_key[:4]}...{api_key[-4:]}'
else:
masked_key = 'not_set_or_too_short'
logger.info('posthog_configuration', extra={'posthog_api_key_masked': masked_key})
# Global toggle for the experiment manager
ENABLE_EXPERIMENT_MANAGER = (
os.environ.get('ENABLE_EXPERIMENT_MANAGER', 'false').lower() == 'true'
)
# Get the current experiment type from environment variable
# If None, no experiment is running
EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT = os.environ.get(
'EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT', ''
)
# System prompt experiment toggle
EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT = os.environ.get(
'EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT', ''
)
EXPERIMENT_CLAUDE4_VS_GPT5 = os.environ.get('EXPERIMENT_CLAUDE4_VS_GPT5', '')
EXPERIMENT_CONDENSER_MAX_STEP = os.environ.get('EXPERIMENT_CONDENSER_MAX_STEP', '')
logger.info(
'experiment_manager:run_conversation_variant_test:experiment_config',
extra={
'enable_experiment_manager': ENABLE_EXPERIMENT_MANAGER,
'experiment_litellm_default_model_experiment': EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT,
'experiment_system_prompt_experiment': EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
'experiment_claude4_vs_gpt5_experiment': EXPERIMENT_CLAUDE4_VS_GPT5,
'experiment_condenser_max_step': EXPERIMENT_CONDENSER_MAX_STEP,
},
)

View File

@@ -1,97 +0,0 @@
from uuid import UUID
from experiments.constants import (
ENABLE_EXPERIMENT_MANAGER,
EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
)
from experiments.experiment_versions import (
handle_system_prompt_experiment,
)
from openhands.core.config.openhands_config import OpenHandsConfig
from openhands.core.logger import openhands_logger as logger
from openhands.experiments.experiment_manager import ExperimentManager
from openhands.sdk import Agent
from openhands.server.session.conversation_init_data import ConversationInitData
class SaaSExperimentManager(ExperimentManager):
@staticmethod
def run_agent_variant_tests__v1(
user_id: str | None, conversation_id: UUID, agent: Agent
) -> Agent:
if not ENABLE_EXPERIMENT_MANAGER:
logger.info(
'experiment_manager:run_conversation_variant_test:skipped',
extra={'reason': 'experiment_manager_disabled'},
)
return agent
if EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT:
agent = agent.model_copy(
update={'system_prompt_filename': 'system_prompt_long_horizon.j2'}
)
return agent
@staticmethod
def run_conversation_variant_test(
user_id, conversation_id, conversation_settings
) -> ConversationInitData:
"""
Run conversation variant test and potentially modify the conversation settings
based on the PostHog feature flags.
Args:
user_id: The user ID
conversation_id: The conversation ID
conversation_settings: The conversation settings that may include convo_id and llm_model
Returns:
The modified conversation settings
"""
logger.debug(
'experiment_manager:run_conversation_variant_test:started',
extra={'user_id': user_id, 'conversation_id': conversation_id},
)
return conversation_settings
@staticmethod
def run_config_variant_test(
user_id: str | None, conversation_id: str, config: OpenHandsConfig
) -> OpenHandsConfig:
"""
Run agent config variant test and potentially modify the OpenHands config
based on the current experiment type and PostHog feature flags.
Args:
user_id: The user ID
conversation_id: The conversation ID
config: The OpenHands configuration
Returns:
The modified OpenHands configuration
"""
logger.info(
'experiment_manager:run_config_variant_test:started',
extra={'user_id': user_id},
)
# Skip all experiment processing if the experiment manager is disabled
if not ENABLE_EXPERIMENT_MANAGER:
logger.info(
'experiment_manager:run_config_variant_test:skipped',
extra={'reason': 'experiment_manager_disabled'},
)
return config
# Pass the entire OpenHands config to the system prompt experiment
# Let the experiment handler directly modify the config as needed
modified_config = handle_system_prompt_experiment(
user_id, conversation_id, config
)
# Condenser max step experiment is applied via conversation variant test,
# not config variant test. Return modified config from system prompt only.
return modified_config

View File

@@ -1,107 +0,0 @@
"""
LiteLLM model experiment handler.
This module contains the handler for the LiteLLM model experiment.
"""
import posthog
from experiments.constants import EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT
from server.constants import (
IS_FEATURE_ENV,
build_litellm_proxy_model_path,
get_default_litellm_model,
)
from openhands.core.logger import openhands_logger as logger
def handle_litellm_default_model_experiment(
user_id, conversation_id, conversation_settings
):
"""
Handle the LiteLLM model experiment.
Args:
user_id: The user ID
conversation_id: The conversation ID
conversation_settings: The conversation settings
Returns:
Modified conversation settings
"""
# No-op if the specific experiment is not enabled
if not EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT:
logger.info(
'experiment_manager:ab_testing:skipped',
extra={
'convo_id': conversation_id,
'reason': 'experiment_not_enabled',
'experiment': EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT,
},
)
return conversation_settings
# Use experiment name as the flag key
try:
enabled_variant = posthog.get_feature_flag(
EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT, conversation_id
)
except Exception as e:
logger.error(
'experiment_manager:get_feature_flag:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT,
'error': str(e),
},
)
return conversation_settings
# Log the experiment event
# If this is a feature environment, add "FEATURE_" prefix to user_id for PostHog
posthog_user_id = f'FEATURE_{user_id}' if IS_FEATURE_ENV else user_id
try:
posthog.capture(
distinct_id=posthog_user_id,
event='model_set',
properties={
'conversation_id': conversation_id,
'variant': enabled_variant,
'original_user_id': user_id,
'is_feature_env': IS_FEATURE_ENV,
},
)
except Exception as e:
logger.error(
'experiment_manager:posthog_capture:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_LITELLM_DEFAULT_MODEL_EXPERIMENT,
'error': str(e),
},
)
# Continue execution as this is not critical
logger.info(
'posthog_capture',
extra={
'event': 'model_set',
'posthog_user_id': posthog_user_id,
'is_feature_env': IS_FEATURE_ENV,
'conversation_id': conversation_id,
'variant': enabled_variant,
},
)
# Set the model based on the feature flag variant
if enabled_variant == 'claude37':
# Use the shared utility to construct the LiteLLM proxy model path
model = build_litellm_proxy_model_path('claude-3-7-sonnet-20250219')
# Update the conversation settings with the selected model
conversation_settings.llm_model = model
else:
# Update the conversation settings with the default model for the current version
conversation_settings.llm_model = get_default_litellm_model()
return conversation_settings

View File

@@ -1,181 +0,0 @@
"""
System prompt experiment handler.
This module contains the handler for the system prompt experiment that uses
the PostHog variant as the system prompt filename.
"""
import copy
import posthog
from experiments.constants import EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT
from server.constants import IS_FEATURE_ENV
from storage.experiment_assignment_store import ExperimentAssignmentStore
from openhands.core.config.openhands_config import OpenHandsConfig
from openhands.core.logger import openhands_logger as logger
def _get_system_prompt_variant(user_id, conversation_id):
"""
Get the system prompt variant for the experiment.
Args:
user_id: The user ID
conversation_id: The conversation ID
Returns:
str or None: The PostHog variant name or None if experiment is not enabled or error occurs
"""
# No-op if the specific experiment is not enabled
if not EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT:
logger.info(
'experiment_manager_002:ab_testing:skipped',
extra={
'convo_id': conversation_id,
'reason': 'experiment_not_enabled',
'experiment': EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
},
)
return None
# Use experiment name as the flag key
try:
enabled_variant = posthog.get_feature_flag(
EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT, conversation_id
)
except Exception as e:
logger.error(
'experiment_manager:get_feature_flag:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
'error': str(e),
},
)
return None
# Store the experiment assignment in the database
try:
experiment_store = ExperimentAssignmentStore()
experiment_store.update_experiment_variant(
conversation_id=conversation_id,
experiment_name='system_prompt_experiment',
variant=enabled_variant,
)
except Exception as e:
logger.error(
'experiment_manager:store_assignment:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
'variant': enabled_variant,
'error': str(e),
},
)
# Fail the experiment if we cannot track the splits - results would not be explainable
return None
# Log the experiment event
# If this is a feature environment, add "FEATURE_" prefix to user_id for PostHog
posthog_user_id = f'FEATURE_{user_id}' if IS_FEATURE_ENV else user_id
try:
posthog.capture(
distinct_id=posthog_user_id,
event='system_prompt_set',
properties={
'conversation_id': conversation_id,
'variant': enabled_variant,
'original_user_id': user_id,
'is_feature_env': IS_FEATURE_ENV,
},
)
except Exception as e:
logger.error(
'experiment_manager:posthog_capture:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
'error': str(e),
},
)
# Continue execution as this is not critical
logger.info(
'posthog_capture',
extra={
'event': 'system_prompt_set',
'posthog_user_id': posthog_user_id,
'is_feature_env': IS_FEATURE_ENV,
'conversation_id': conversation_id,
'variant': enabled_variant,
},
)
return enabled_variant
def handle_system_prompt_experiment(
user_id, conversation_id, config: OpenHandsConfig
) -> OpenHandsConfig:
"""
Handle the system prompt experiment for OpenHands config.
Args:
user_id: The user ID
conversation_id: The conversation ID
config: The OpenHands configuration
Returns:
Modified OpenHands configuration
"""
enabled_variant = _get_system_prompt_variant(user_id, conversation_id)
# If variant is None, experiment is not enabled or there was an error
if enabled_variant is None:
return config
# Deep copy the config to avoid modifying the original
modified_config = copy.deepcopy(config)
# Set the system prompt filename based on the variant
if enabled_variant == 'control':
# Use the long-horizon system prompt for the control variant
agent_config = modified_config.get_agent_config(modified_config.default_agent)
agent_config.system_prompt_filename = 'system_prompt_long_horizon.j2'
agent_config.enable_plan_mode = True
elif enabled_variant == 'interactive':
modified_config.get_agent_config(
modified_config.default_agent
).system_prompt_filename = 'system_prompt_interactive.j2'
elif enabled_variant == 'no_tools':
modified_config.get_agent_config(
modified_config.default_agent
).system_prompt_filename = 'system_prompt.j2'
else:
logger.error(
'system_prompt_experiment:unknown_variant',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'reason': 'no explicit mapping; returning original config',
},
)
return config
# Log which prompt is being used
logger.info(
'system_prompt_experiment:prompt_selected',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'system_prompt_filename': modified_config.get_agent_config(
modified_config.default_agent
).system_prompt_filename,
'variant': enabled_variant,
},
)
return modified_config

View File

@@ -1,137 +0,0 @@
"""
LiteLLM model experiment handler.
This module contains the handler for the LiteLLM model experiment.
"""
import posthog
from experiments.constants import EXPERIMENT_CLAUDE4_VS_GPT5
from server.constants import (
IS_FEATURE_ENV,
build_litellm_proxy_model_path,
get_default_litellm_model,
)
from storage.experiment_assignment_store import ExperimentAssignmentStore
from openhands.core.logger import openhands_logger as logger
from openhands.server.session.conversation_init_data import ConversationInitData
def _get_model_variant(user_id: str | None, conversation_id: str) -> str | None:
if not EXPERIMENT_CLAUDE4_VS_GPT5:
logger.info(
'experiment_manager:ab_testing:skipped',
extra={
'convo_id': conversation_id,
'reason': 'experiment_not_enabled',
'experiment': EXPERIMENT_CLAUDE4_VS_GPT5,
},
)
return None
try:
enabled_variant = posthog.get_feature_flag(
EXPERIMENT_CLAUDE4_VS_GPT5, conversation_id
)
except Exception as e:
logger.error(
'experiment_manager:get_feature_flag:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CLAUDE4_VS_GPT5,
'error': str(e),
},
)
return None
# Store the experiment assignment in the database
try:
experiment_store = ExperimentAssignmentStore()
experiment_store.update_experiment_variant(
conversation_id=conversation_id,
experiment_name='claude4_vs_gpt5_experiment',
variant=enabled_variant,
)
except Exception as e:
logger.error(
'experiment_manager:store_assignment:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CLAUDE4_VS_GPT5,
'variant': enabled_variant,
'error': str(e),
},
)
# Fail the experiment if we cannot track the splits - results would not be explainable
return None
# Log the experiment event
# If this is a feature environment, add "FEATURE_" prefix to user_id for PostHog
posthog_user_id = f'FEATURE_{user_id}' if IS_FEATURE_ENV else user_id
try:
posthog.capture(
distinct_id=posthog_user_id,
event='claude4_or_gpt5_set',
properties={
'conversation_id': conversation_id,
'variant': enabled_variant,
'original_user_id': user_id,
'is_feature_env': IS_FEATURE_ENV,
},
)
except Exception as e:
logger.error(
'experiment_manager:posthog_capture:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CLAUDE4_VS_GPT5,
'error': str(e),
},
)
# Continue execution as this is not critical
logger.info(
'posthog_capture',
extra={
'event': 'claude4_or_gpt5_set',
'posthog_user_id': posthog_user_id,
'is_feature_env': IS_FEATURE_ENV,
'conversation_id': conversation_id,
'variant': enabled_variant,
},
)
return enabled_variant
def handle_claude4_vs_gpt5_experiment(
user_id: str | None,
conversation_id: str,
conversation_settings: ConversationInitData,
) -> ConversationInitData:
"""
Handle the LiteLLM model experiment.
Args:
user_id: The user ID
conversation_id: The conversation ID
conversation_settings: The conversation settings
Returns:
Modified conversation settings
"""
enabled_variant = _get_model_variant(user_id, conversation_id)
if not enabled_variant:
return conversation_settings
# Set the model based on the feature flag variant
if enabled_variant == 'gpt5':
model = build_litellm_proxy_model_path('gpt-5-2025-08-07')
conversation_settings.llm_model = model
else:
conversation_settings.llm_model = get_default_litellm_model()
return conversation_settings

View File

@@ -1,232 +0,0 @@
"""
Condenser max step experiment handler.
This module contains the handler for the condenser max step experiment that tests
different max_size values for the condenser configuration.
"""
from uuid import UUID
import posthog
from experiments.constants import EXPERIMENT_CONDENSER_MAX_STEP
from server.constants import IS_FEATURE_ENV
from storage.experiment_assignment_store import ExperimentAssignmentStore
from openhands.core.logger import openhands_logger as logger
from openhands.sdk import Agent
from openhands.sdk.context.condenser import (
LLMSummarizingCondenser,
)
from openhands.server.session.conversation_init_data import ConversationInitData
def _get_condenser_max_step_variant(user_id, conversation_id):
"""
Get the condenser max step variant for the experiment.
Args:
user_id: The user ID
conversation_id: The conversation ID
Returns:
str or None: The PostHog variant name or None if experiment is not enabled or error occurs
"""
# No-op if the specific experiment is not enabled
if not EXPERIMENT_CONDENSER_MAX_STEP:
logger.info(
'experiment_manager_004:ab_testing:skipped',
extra={
'convo_id': conversation_id,
'reason': 'experiment_not_enabled',
'experiment': EXPERIMENT_CONDENSER_MAX_STEP,
},
)
return None
# Use experiment name as the flag key
try:
enabled_variant = posthog.get_feature_flag(
EXPERIMENT_CONDENSER_MAX_STEP, conversation_id
)
except Exception as e:
logger.error(
'experiment_manager:get_feature_flag:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CONDENSER_MAX_STEP,
'error': str(e),
},
)
return None
# Store the experiment assignment in the database
try:
experiment_store = ExperimentAssignmentStore()
experiment_store.update_experiment_variant(
conversation_id=conversation_id,
experiment_name='condenser_max_step_experiment',
variant=enabled_variant,
)
except Exception as e:
logger.error(
'experiment_manager:store_assignment:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CONDENSER_MAX_STEP,
'variant': enabled_variant,
'error': str(e),
},
)
# Fail the experiment if we cannot track the splits - results would not be explainable
return None
# Log the experiment event
# If this is a feature environment, add "FEATURE_" prefix to user_id for PostHog
posthog_user_id = f'FEATURE_{user_id}' if IS_FEATURE_ENV else user_id
try:
posthog.capture(
distinct_id=posthog_user_id,
event='condenser_max_step_set',
properties={
'conversation_id': conversation_id,
'variant': enabled_variant,
'original_user_id': user_id,
'is_feature_env': IS_FEATURE_ENV,
},
)
except Exception as e:
logger.error(
'experiment_manager:posthog_capture:failed',
extra={
'convo_id': conversation_id,
'experiment': EXPERIMENT_CONDENSER_MAX_STEP,
'error': str(e),
},
)
# Continue execution as this is not critical
logger.info(
'posthog_capture',
extra={
'event': 'condenser_max_step_set',
'posthog_user_id': posthog_user_id,
'is_feature_env': IS_FEATURE_ENV,
'conversation_id': conversation_id,
'variant': enabled_variant,
},
)
return enabled_variant
def handle_condenser_max_step_experiment(
user_id: str | None,
conversation_id: str,
conversation_settings: ConversationInitData,
) -> ConversationInitData:
"""
Handle the condenser max step experiment for conversation settings.
We should not modify persistent user settings. Instead, apply the experiment
variant to the conversation's in-memory settings object for this session only.
Variants:
- control -> condenser_max_size = 120
- treatment -> condenser_max_size = 80
Returns the (potentially) modified conversation_settings.
"""
enabled_variant = _get_condenser_max_step_variant(user_id, conversation_id)
if enabled_variant is None:
return conversation_settings
if enabled_variant == 'control':
condenser_max_size = 120
elif enabled_variant == 'treatment':
condenser_max_size = 80
else:
logger.error(
'condenser_max_step_experiment:unknown_variant',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'reason': 'unknown variant; returning original conversation settings',
},
)
return conversation_settings
try:
# Apply the variant to this conversation only; do not persist to DB.
# Not all OpenHands versions expose `condenser_max_size` on settings.
if hasattr(conversation_settings, 'condenser_max_size'):
conversation_settings.condenser_max_size = condenser_max_size
logger.info(
'condenser_max_step_experiment:conversation_settings_applied',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'condenser_max_size': condenser_max_size,
},
)
else:
logger.warning(
'condenser_max_step_experiment:field_missing_on_settings',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'reason': 'condenser_max_size not present on ConversationInitData',
},
)
except Exception as e:
logger.error(
'condenser_max_step_experiment:apply_failed',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'error': str(e),
},
)
return conversation_settings
return conversation_settings
def handle_condenser_max_step_experiment__v1(
user_id: str | None,
conversation_id: UUID,
agent: Agent,
) -> Agent:
enabled_variant = _get_condenser_max_step_variant(user_id, str(conversation_id))
if enabled_variant is None:
return agent
if enabled_variant == 'control':
condenser_max_size = 120
elif enabled_variant == 'treatment':
condenser_max_size = 80
else:
logger.error(
'condenser_max_step_experiment:unknown_variant',
extra={
'user_id': user_id,
'convo_id': conversation_id,
'variant': enabled_variant,
'reason': 'unknown variant; returning original conversation settings',
},
)
return agent
condenser_llm = agent.llm.model_copy(update={'usage_id': 'condenser'})
condenser = LLMSummarizingCondenser(
llm=condenser_llm, max_size=condenser_max_size, keep_first=4
)
return agent.model_copy(update={'condenser': condenser})

View File

@@ -1,25 +0,0 @@
"""
Experiment versions package.
This package contains handlers for different experiment versions.
"""
from experiments.experiment_versions._001_litellm_default_model_experiment import (
handle_litellm_default_model_experiment,
)
from experiments.experiment_versions._002_system_prompt_experiment import (
handle_system_prompt_experiment,
)
from experiments.experiment_versions._003_llm_claude4_vs_gpt5_experiment import (
handle_claude4_vs_gpt5_experiment,
)
from experiments.experiment_versions._004_condenser_max_step_experiment import (
handle_condenser_max_step_experiment,
)
__all__ = [
'handle_litellm_default_model_experiment',
'handle_system_prompt_experiment',
'handle_claude4_vs_gpt5_experiment',
'handle_condenser_max_step_experiment',
]

View File

@@ -0,0 +1,65 @@
from pydantic import SecretStr
from server.auth.token_manager import TokenManager
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.bitbucket_data_center.bitbucket_dc_service import (
BitbucketDCService,
)
from openhands.integrations.service_types import ProviderType
class SaaSBitbucketDCService(BitbucketDCService):
def __init__(
self,
user_id: str | None = None,
external_auth_token: SecretStr | None = None,
external_auth_id: str | None = None,
token: SecretStr | None = None,
external_token_manager: bool = False,
base_domain: str | None = None,
):
logger.debug(
f'SaaSBitbucketDCService created with user_id {user_id}, external_auth_id {external_auth_id}, external_auth_token {'set' if external_auth_token else 'None'}, token {'set' if token else 'None'}, external_token_manager {external_token_manager}'
)
super().__init__(
user_id=user_id,
external_auth_token=external_auth_token,
external_auth_id=external_auth_id,
token=token,
external_token_manager=external_token_manager,
base_domain=base_domain,
)
self.token_manager = TokenManager(external=external_token_manager)
self.refresh = True
async def get_latest_token(self) -> SecretStr | None:
bitbucket_dc_token = None
if self.external_auth_token:
bitbucket_dc_token = SecretStr(
await self.token_manager.get_idp_token(
self.external_auth_token.get_secret_value(),
idp=ProviderType.BITBUCKET_DATA_CENTER,
)
)
logger.debug('Got Bitbucket DC token via external_auth_token')
elif self.external_auth_id:
offline_token = await self.token_manager.load_offline_token(
self.external_auth_id
)
bitbucket_dc_token = SecretStr(
await self.token_manager.get_idp_token_from_offline_token(
offline_token, ProviderType.BITBUCKET_DATA_CENTER
)
)
logger.debug('Got Bitbucket DC token via external_auth_id')
elif self.user_id:
bitbucket_dc_token = SecretStr(
await self.token_manager.get_idp_token_from_idp_user_id(
self.user_id, ProviderType.BITBUCKET_DATA_CENTER
)
)
logger.debug('Got Bitbucket DC token via user_id')
else:
logger.warning('external_auth_token and user_id not set!')
return bitbucket_dc_token

View File

@@ -6,7 +6,7 @@ from datetime import datetime
from enum import Enum
from typing import Any
from github import Github, GithubIntegration
from github import Auth, Github, GithubIntegration
from integrations.github.github_view import (
GithubIssue,
)
@@ -84,7 +84,7 @@ class GitHubDataCollector:
# self.full_saved_pr_path = 'github_data/prs/{}-{}/data.json'
self.full_saved_pr_path = 'prs/github/{}-{}/data.json'
self.github_integration = GithubIntegration(
GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
auth=Auth.AppAuth(GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY)
)
self.conversation_id = None
@@ -116,10 +116,8 @@ class GitHubDataCollector:
return suffix
def _get_installation_access_token(self, installation_id: str) -> str:
token_data = self.github_integration.get_access_token(
installation_id # type: ignore[arg-type]
)
def _get_installation_access_token(self, installation_id: int) -> str:
token_data = self.github_integration.get_access_token(installation_id)
return token_data.token
def _check_openhands_author(self, name, login) -> bool:
@@ -134,7 +132,7 @@ class GitHubDataCollector:
)
def _get_issue_comments(
self, installation_id: str, repo_name: str, issue_number: int, conversation_id
self, installation_id: int, repo_name: str, issue_number: int, conversation_id
) -> list[dict[str, Any]]:
"""
Retrieve all comments from an issue until a comment with conversation_id is found
@@ -143,7 +141,7 @@ class GitHubDataCollector:
try:
installation_token = self._get_installation_access_token(installation_id)
with Github(installation_token) as github_client:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(repo_name)
issue = repo.get_issue(issue_number)
comments = []
@@ -234,10 +232,10 @@ class GitHubDataCollector:
f'[Github]: Saved issue #{issue_number} for {github_view.full_repo_name}'
)
def _get_pr_commits(self, installation_id: str, repo_name: str, pr_number: int):
def _get_pr_commits(self, installation_id: int, repo_name: str, pr_number: int):
commits = []
installation_token = self._get_installation_access_token(installation_id)
with Github(installation_token) as github_client:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(repo_name)
pr = repo.get_pull(pr_number)
@@ -431,7 +429,7 @@ class GitHubDataCollector:
- Num openhands review comments
"""
pr_number = openhands_pr.pr_number
installation_id = openhands_pr.installation_id
installation_id = int(openhands_pr.installation_id)
repo_id = openhands_pr.repo_id
# Get installation token and create Github client
@@ -569,7 +567,7 @@ class GitHubDataCollector:
openhands_helped_author = openhands_commit_count > 0
# Update the PR with OpenHands statistics
update_success = store.update_pr_openhands_stats(
update_success = await store.update_pr_openhands_stats(
repo_id=repo_id,
pr_number=pr_number,
original_updated_at=openhands_pr.updated_at,
@@ -612,7 +610,7 @@ class GitHubDataCollector:
action = payload.get('action', '')
return action == 'closed' and 'pull_request' in payload
def _track_closed_or_merged_pr(self, payload):
async def _track_closed_or_merged_pr(self, payload):
"""
Track PR closed/merged event
"""
@@ -671,17 +669,17 @@ class GitHubDataCollector:
num_general_comments=num_general_comments,
)
store.insert_pr(pr)
await store.insert_pr(pr)
logger.info(f'Tracked PR {status}: {repo_id}#{pr_number}')
def process_payload(self, message: Message):
async def process_payload(self, message: Message):
if not COLLECT_GITHUB_INTERACTIONS:
return
raw_payload = message.message.get('payload', {})
if self._is_pr_closed_or_merged(raw_payload):
self._track_closed_or_merged_pr(raw_payload)
await self._track_closed_or_merged_pr(raw_payload)
async def save_data(self, github_view: ResolverViewInterface):
if not COLLECT_GITHUB_INTERACTIONS:

View File

@@ -1,6 +1,6 @@
from types import MappingProxyType
from github import Github, GithubIntegration
from github import Auth, Github, GithubIntegration
from integrations.github.data_collector import GitHubDataCollector
from integrations.github.github_solvability import summarize_issue_solvability
from integrations.github.github_view import (
@@ -10,6 +10,7 @@ from integrations.github.github_view import (
GithubIssue,
GithubIssueComment,
GithubPRComment,
GithubViewType,
)
from integrations.manager import Manager
from integrations.models import (
@@ -19,31 +20,39 @@ from integrations.models import (
from integrations.types import ResolverViewInterface
from integrations.utils import (
CONVERSATION_URL,
ENABLE_SOLVABILITY_ANALYSIS,
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
get_session_expired_message,
get_user_not_found_message,
)
from integrations.v1_utils import get_saas_user_auth
from jinja2 import Environment, FileSystemLoader
from pydantic import SecretStr
from server.auth.auth_error import ExpiredError
from server.auth.constants import GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
from server.auth.token_manager import TokenManager
from server.utils.conversation_callback_utils import register_callback_processor
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.provider import ProviderToken, ProviderType
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.integrations.service_types import AuthenticationError
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.storage.data_models.secrets import Secrets
from openhands.utils.async_utils import call_sync_from_async
class GithubManager(Manager):
class GithubManager(Manager[GithubViewType]):
def __init__(
self, token_manager: TokenManager, data_collector: GitHubDataCollector
):
self.token_manager = token_manager
self.data_collector = data_collector
self.github_integration = GithubIntegration(
GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
auth=Auth.AppAuth(GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY)
)
self.jinja_env = Environment(
@@ -60,11 +69,8 @@ class GithubManager(Manager):
return f'{owner}/{repo_name}'
def _get_installation_access_token(self, installation_id: str) -> str:
# get_access_token is typed to only accept int, but it can handle str.
token_data = self.github_integration.get_access_token(
installation_id # type: ignore[arg-type]
)
def _get_installation_access_token(self, installation_id: int) -> str:
token_data = self.github_integration.get_access_token(installation_id)
return token_data.token
def _add_reaction(
@@ -77,7 +83,7 @@ class GithubManager(Manager):
reaction: The reaction to add (e.g. "eyes", "+1", "-1", "laugh", "confused", "heart", "hooray", "rocket")
installation_token: GitHub installation access token for API access
"""
with Github(installation_token) as github_client:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(github_view.full_repo_name)
# Add reaction based on view type
if isinstance(github_view, GithubInlinePRComment):
@@ -119,6 +125,76 @@ class GithubManager(Manager):
return False
def _get_issue_number_from_payload(self, message: Message) -> int | None:
"""Extract issue/PR number from a GitHub webhook payload.
Supports all event types that can trigger jobs:
- Labeled issues: payload['issue']['number']
- Issue comments: payload['issue']['number']
- PR comments: payload['issue']['number'] (PRs are accessed via issue endpoint)
- Inline PR comments: payload['pull_request']['number']
Args:
message: The incoming GitHub webhook message
Returns:
The issue/PR number, or None if not found
"""
payload = message.message.get('payload', {})
# Labeled issues, issue comments, and PR comments all have 'issue' in payload
if 'issue' in payload:
return payload['issue']['number']
# Inline PR comments have 'pull_request' directly in payload
if 'pull_request' in payload:
return payload['pull_request']['number']
return None
def _send_user_not_found_message(self, message: Message, username: str):
"""Send a message to the user informing them they need to create an OpenHands account.
This method handles all supported trigger types:
- Labeled issues (action='labeled' with openhands label)
- Issue comments (comment containing @openhands)
- PR comments (comment containing @openhands on a PR)
- Inline PR review comments (comment containing @openhands)
Args:
message: The incoming GitHub webhook message
username: The GitHub username to mention in the response
"""
payload = message.message.get('payload', {})
installation_id = message.message['installation']
repo_obj = payload['repository']
full_repo_name = self._get_full_repo_name(repo_obj)
# Get installation token to post the comment
installation_token = self._get_installation_access_token(installation_id)
# Determine the issue/PR number based on the event type
issue_number = self._get_issue_number_from_payload(message)
if not issue_number:
logger.warning(
f'[GitHub] Could not determine issue/PR number to send user not found message for {username}. '
f'Payload keys: {list(payload.keys())}'
)
return
# Post the comment
try:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(full_repo_name)
issue = repo.get_issue(number=issue_number)
issue.create_comment(get_user_not_found_message(username))
except Exception as e:
logger.error(
f'[GitHub] Failed to send user not found message to {username} '
f'on {full_repo_name}#{issue_number}: {e}'
)
async def is_job_requested(self, message: Message) -> bool:
self._confirm_incoming_source_type(message)
@@ -138,11 +214,7 @@ class GithubManager(Manager):
).get('body', ''):
return False
if GithubFactory.is_eligible_for_conversation_starter(
message
) and self._user_has_write_access_to_repo(installation_id, repo_name, username):
await GithubFactory.trigger_conversation_starter(message)
# Check event types before making expensive API calls (e.g., _user_has_write_access_to_repo)
if not (
GithubFactory.is_labeled_issue(message)
or GithubFactory.is_issue_comment(message)
@@ -152,13 +224,22 @@ class GithubManager(Manager):
return False
logger.info(f'[GitHub] Checking permissions for {username} in {repo_name}')
user_has_write_access = self._user_has_write_access_to_repo(
installation_id, repo_name, username
)
return self._user_has_write_access_to_repo(installation_id, repo_name, username)
if (
GithubFactory.is_eligible_for_conversation_starter(message)
and user_has_write_access
):
await GithubFactory.trigger_conversation_starter(message)
return user_has_write_access
async def receive_message(self, message: Message):
self._confirm_incoming_source_type(message)
try:
await call_sync_from_async(self.data_collector.process_payload, message)
await self.data_collector.process_payload(message)
except Exception:
logger.warning(
'[Github]: Error processing payload for gh interaction', exc_info=True
@@ -167,9 +248,20 @@ class GithubManager(Manager):
if await self.is_job_requested(message):
payload = message.message.get('payload', {})
user_id = payload['sender']['id']
username = payload['sender']['login']
keycloak_user_id = await self.token_manager.get_user_id_from_idp_user_id(
user_id, ProviderType.GITHUB
)
# Check if the user has an OpenHands account
if not keycloak_user_id:
logger.warning(
f'[GitHub] User {username} (id={user_id}) not found in Keycloak. '
f'User must create an OpenHands account first.'
)
self._send_user_not_found_message(message, username)
return
github_view = await GithubFactory.create_github_view_from_payload(
message, keycloak_user_id
)
@@ -181,46 +273,51 @@ class GithubManager(Manager):
github_view.installation_id
)
# Store the installation token
self.token_manager.store_org_token(
await self.token_manager.store_org_token(
github_view.installation_id, installation_token
)
# Add eyes reaction to acknowledge we've read the request
self._add_reaction(github_view, 'eyes', installation_token)
await self.start_job(github_view)
async def send_message(self, message: Message, github_view: ResolverViewInterface):
installation_token = self.token_manager.load_org_token(
async def send_message(self, message: str, github_view: GithubViewType):
"""Send a message to GitHub.
Args:
message: The message content to send (plain text string)
github_view: The GitHub view object containing issue/PR/comment info
"""
installation_token = await self.token_manager.load_org_token(
github_view.installation_id
)
if not installation_token:
logger.warning('Missing installation token')
return
outgoing_message = message.message
if isinstance(github_view, GithubInlinePRComment):
with Github(installation_token) as github_client:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(github_view.full_repo_name)
pr = repo.get_pull(github_view.issue_number)
pr.create_review_comment_reply(
comment_id=github_view.comment_id, body=outgoing_message
comment_id=github_view.comment_id, body=message
)
elif (
isinstance(github_view, GithubPRComment)
or isinstance(github_view, GithubIssueComment)
or isinstance(github_view, GithubIssue)
elif isinstance(
github_view, (GithubPRComment, GithubIssueComment, GithubIssue)
):
with Github(installation_token) as github_client:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(github_view.full_repo_name)
issue = repo.get_issue(number=github_view.issue_number)
issue.create_comment(outgoing_message)
issue.create_comment(message)
else:
logger.warning('Unsupported location')
# Catch any new types added to GithubViewType that aren't handled above
logger.warning( # type: ignore[unreachable]
f'Unsupported github_view type: {type(github_view).__name__}'
)
return
async def start_job(self, github_view: ResolverViewInterface):
async def start_job(self, github_view: GithubViewType) -> None:
"""Kick off a job with openhands agent.
1. Get user credential
@@ -233,7 +330,7 @@ class GithubManager(Manager):
)
try:
msg_info = None
msg_info: str = ''
try:
user_info = github_view.user_info
@@ -274,19 +371,19 @@ class GithubManager(Manager):
# 3. Once the conversation is started, its base cost will include the report's spend as well which allows us to control max budget per resolver task
convo_metadata = await github_view.initialize_new_conversation()
solvability_summary = None
try:
if user_token:
if not ENABLE_SOLVABILITY_ANALYSIS:
logger.info(
'[Github]: Solvability report feature is disabled, skipping'
)
else:
try:
solvability_summary = await summarize_issue_solvability(
github_view, user_token
)
else:
except Exception as e:
logger.warning(
'[Github]: No user token available for solvability analysis'
f'[Github]: Error summarizing issue solvability: {str(e)}'
)
except Exception as e:
logger.warning(
f'[Github]: Error summarizing issue solvability: {str(e)}'
)
saas_user_auth = await get_saas_user_auth(
github_view.user_info.keycloak_user_id, self.token_manager
@@ -305,7 +402,7 @@ class GithubManager(Manager):
f'[GitHub] Created conversation {conversation_id} for user {user_info.username}'
)
if not github_view.v1:
if not github_view.v1_enabled:
# Create a GithubCallbackProcessor
processor = GithubCallbackProcessor(
github_view=github_view,
@@ -342,15 +439,20 @@ class GithubManager(Manager):
msg_info = f'@{user_info.username} please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
msg = self.create_outgoing_message(msg_info)
await self.send_message(msg, github_view)
except (AuthenticationError, ExpiredError, SessionExpiredError) as e:
logger.warning(
f'[GitHub] Session expired for user {user_info.username}: {str(e)}'
)
msg_info = get_session_expired_message(user_info.username)
await self.send_message(msg_info, github_view)
except Exception:
logger.exception('[Github]: Error starting job')
msg = self.create_outgoing_message(
msg='Uh oh! There was an unexpected error starting the job :('
await self.send_message(
'Uh oh! There was an unexpected error starting the job :(', github_view
)
await self.send_message(msg, github_view)
try:
await self.data_collector.save_data(github_view)

View File

@@ -1,6 +1,6 @@
import asyncio
from integrations.utils import store_repositories_in_db
from integrations.store_repo_utils import store_repositories_in_db
from pydantic import SecretStr
from server.auth.token_manager import TokenManager
@@ -122,13 +122,37 @@ class SaaSGitHubService(GitHubService):
raise Exception(f'No node_id found for repository {repo_id}')
return node_id
async def _get_external_auth_id(self) -> str | None:
"""Get or fetch external_auth_id from Keycloak token if not already set."""
if self.external_auth_id:
return self.external_auth_id
if self.external_auth_token:
try:
user_info = await self.token_manager.get_user_info(
self.external_auth_token.get_secret_value()
)
self.external_auth_id = user_info.sub
logger.info(
f'Determined external_auth_id from Keycloak token: {self.external_auth_id}'
)
return self.external_auth_id
except Exception as e:
logger.warning(
f'Could not determine external_auth_id from token: {e}',
exc_info=True,
)
return None
async def get_paginated_repos(self, page, per_page, sort, installation_id):
repositories = await super().get_paginated_repos(
page, per_page, sort, installation_id
)
asyncio.create_task(
store_repositories_in_db(repositories, self.external_auth_id)
)
external_auth_id = await self._get_external_auth_id()
if external_auth_id:
asyncio.create_task(
store_repositories_in_db(repositories, external_auth_id)
)
return repositories
async def get_all_repositories(
@@ -136,8 +160,10 @@ class SaaSGitHubService(GitHubService):
) -> list[Repository]:
repositories = await super().get_all_repositories(sort, app_mode)
# Schedule the background task without awaiting it
asyncio.create_task(
store_repositories_in_db(repositories, self.external_auth_id)
)
external_auth_id = await self._get_external_auth_id()
if external_auth_id:
asyncio.create_task(
store_repositories_in_db(repositories, external_auth_id)
)
# Return repositories immediately
return repositories

View File

@@ -1,7 +1,7 @@
import asyncio
import time
from github import Github
from github import Auth, Github
from integrations.github.github_view import (
GithubInlinePRComment,
GithubIssueComment,
@@ -14,7 +14,6 @@ from integrations.solvability.models.summary import SolvabilitySummary
from integrations.utils import ENABLE_SOLVABILITY_ANALYSIS
from pydantic import ValidationError
from server.config import get_config
from storage.database import session_maker
from storage.saas_settings_store import SaasSettingsStore
from openhands.core.config import LLMConfig
@@ -47,7 +46,7 @@ def fetch_github_issue_context(
context_parts.append(f'Title: {github_view.title}')
context_parts.append(f'Description:\n{github_view.description}')
with Github(user_token) as github_client:
with Github(auth=Auth.Token(user_token)) as github_client:
repo = github_client.get_repo(github_view.full_repo_name)
issue = repo.get_issue(github_view.issue_number)
if issue.labels:
@@ -90,7 +89,6 @@ async def summarize_issue_solvability(
# Grab the user's information so we can load their LLM configuration
store = SaasSettingsStore(
user_id=github_view.user_info.keycloak_user_id,
session_maker=session_maker,
config=get_config(),
)
@@ -108,6 +106,11 @@ async def summarize_issue_solvability(
f'Solvability analysis disabled for user {github_view.user_info.user_id}'
)
if user_settings.llm_api_key is None:
raise ValueError(
f'[Solvability] No LLM API key found for user {github_view.user_info.user_id}'
)
try:
llm_config = LLMConfig(
model=user_settings.llm_model,

View File

@@ -1,11 +1,13 @@
import logging
import os
from typing import Any
from uuid import UUID
import httpx
from github import Github, GithubIntegration
from github import Auth, Github, GithubException, GithubIntegration
from integrations.utils import get_summary_instruction
from integrations.v1_utils import handle_callback_error
from pydantic import Field
from server.auth.constants import GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
from openhands.agent_server.models import AskAgentRequest, AskAgentResponse
from openhands.app_server.event_callback.event_callback_models import (
@@ -20,8 +22,6 @@ from openhands.app_server.event_callback.util import (
ensure_conversation_found,
ensure_running_sandbox,
get_agent_server_url_from_sandbox,
get_conversation_url,
get_prompt_template,
)
from openhands.sdk import Event
from openhands.sdk.event import ConversationStateUpdateEvent
@@ -34,7 +34,6 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
github_view_data: dict[str, Any] = Field(default_factory=dict)
should_request_summary: bool = Field(default=True)
should_extract: bool = Field(default=True)
inline_pr_comment: bool = Field(default=False)
async def __call__(
@@ -44,7 +43,6 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
event: Event,
) -> EventCallbackResult | None:
"""Process events for GitHub V1 integration."""
# Only handle ConversationStateUpdateEvent
if not isinstance(event, ConversationStateUpdateEvent):
return None
@@ -64,7 +62,12 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
self.should_request_summary = False
try:
_logger.info(f'[GitHub V1] Requesting summary {conversation_id}')
summary = await self._request_summary(conversation_id)
_logger.info(
f'[GitHub V1] Posting summary {conversation_id}',
extra={'summary': summary},
)
await self._post_summary_to_github(summary)
return EventCallbackResult(
@@ -75,25 +78,20 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
detail=summary,
)
except Exception as e:
_logger.exception('[GitHub V1] Error processing callback: %s', e)
# Only try to post error to GitHub if we have basic requirements
try:
# Check if we have installation ID and credentials before posting
if (
self.github_view_data.get('installation_id')
and os.getenv('GITHUB_APP_CLIENT_ID')
and os.getenv('GITHUB_APP_PRIVATE_KEY')
):
await self._post_summary_to_github(
f'OpenHands encountered an error: **{str(e)}**.\n\n'
f'[See the conversation]({get_conversation_url().format(conversation_id)})'
'for more information.'
)
except Exception as post_error:
_logger.warning(
'[GitHub V1] Failed to post error message to GitHub: %s', post_error
)
# Check if we have installation ID and credentials before posting
can_post_error = bool(
self.github_view_data.get('installation_id')
and GITHUB_APP_CLIENT_ID
and GITHUB_APP_PRIVATE_KEY
)
await handle_callback_error(
error=e,
conversation_id=conversation_id,
service_name='GitHub',
service_logger=_logger,
can_post_error=can_post_error,
post_error_func=self._post_summary_to_github,
)
return EventCallbackResult(
status=EventCallbackResultStatus.ERROR,
@@ -115,17 +113,11 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
f'Missing installation ID for GitHub payload: {self.github_view_data}'
)
github_app_client_id = os.getenv('GITHUB_APP_CLIENT_ID', '').strip()
github_app_private_key = os.getenv('GITHUB_APP_PRIVATE_KEY', '').replace(
'\\n', '\n'
)
if not github_app_client_id or not github_app_private_key:
if not GITHUB_APP_CLIENT_ID or not GITHUB_APP_PRIVATE_KEY:
raise ValueError('GitHub App credentials are not configured')
github_integration = GithubIntegration(
github_app_client_id,
github_app_private_key,
auth=Auth.AppAuth(GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY),
)
token_data = github_integration.get_access_token(installation_id)
return token_data.token
@@ -140,19 +132,30 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
full_repo_name = self.github_view_data['full_repo_name']
issue_number = self.github_view_data['issue_number']
if self.inline_pr_comment:
with Github(installation_token) as github_client:
repo = github_client.get_repo(full_repo_name)
pr = repo.get_pull(issue_number)
pr.create_review_comment_reply(
comment_id=self.github_view_data.get('comment_id', ''), body=summary
)
return
try:
if self.inline_pr_comment:
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(full_repo_name)
pr = repo.get_pull(issue_number)
pr.create_review_comment_reply(
comment_id=self.github_view_data.get('comment_id', ''),
body=summary,
)
return
with Github(installation_token) as github_client:
repo = github_client.get_repo(full_repo_name)
issue = repo.get_issue(number=issue_number)
issue.create_comment(summary)
with Github(auth=Auth.Token(installation_token)) as github_client:
repo = github_client.get_repo(full_repo_name)
issue = repo.get_issue(number=issue_number)
issue.create_comment(summary)
except GithubException as e:
if e.status == 410:
_logger.info(
'[GitHub V1] Issue/PR %s#%s was deleted, skipping summary post',
full_repo_name,
issue_number,
)
else:
raise
# -------------------------------------------------------------------------
# Agent / sandbox helpers
@@ -170,8 +173,8 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
send_message_request = AskAgentRequest(question=message_content)
url = (
f'{agent_server_url.rstrip("/")}'
f'/api/conversations/{conversation_id}/ask_agent'
f"{agent_server_url.rstrip('/')}"
f"/api/conversations/{conversation_id}/ask_agent"
)
headers = {'X-Session-API-Key': session_api_key}
payload = send_message_request.model_dump()
@@ -233,8 +236,7 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
# -------------------------------------------------------------------------
async def _request_summary(self, conversation_id: UUID) -> str:
"""
Ask the agent to produce a summary of its work and return the agent response.
"""Ask the agent to produce a summary of its work and return the agent response.
NOTE: This method now returns a string (the agent server's response text)
and raises exceptions on errors. The wrapping into EventCallbackResult
@@ -275,16 +277,16 @@ class GithubV1CallbackProcessor(EventCallbackProcessor):
app_conversation_info.sandbox_id,
)
assert sandbox.session_api_key is not None, (
f'No session API key for sandbox: {sandbox.id}'
)
assert (
sandbox.session_api_key is not None
), f'No session API key for sandbox: {sandbox.id}'
# 3. URL + instruction
agent_server_url = get_agent_server_url_from_sandbox(sandbox)
agent_server_url = get_agent_server_url_from_sandbox(sandbox)
# Prepare message based on agent state
message_content = get_prompt_template('summary_prompt.j2')
message_content = get_summary_instruction()
# Ask the agent and return the response text
return await self._ask_question(

View File

@@ -1,7 +1,7 @@
from dataclasses import dataclass
from uuid import UUID, uuid4
from github import Github, GithubIntegration
from github import Auth, Github, GithubIntegration
from github.Issue import Issue
from integrations.github.github_types import (
WorkflowRun,
@@ -17,16 +17,16 @@ from integrations.utils import (
HOST,
HOST_URL,
get_oh_labels,
get_user_v1_enabled_setting,
has_exact_mention,
)
from jinja2 import Environment
from server.auth.constants import GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
from server.auth.token_manager import TokenManager
from server.config import get_config
from storage.database import session_maker
from storage.org_store import OrgStore
from storage.proactive_conversation_store import ProactiveConversationStore
from storage.saas_secrets_store import SaasSecretsStore
from storage.saas_settings_store import SaasSettingsStore
from openhands.agent_server.models import SendMessageRequest
from openhands.app_server.app_conversation.app_conversation_models import (
@@ -55,6 +55,10 @@ from openhands.utils.async_utils import call_sync_from_async
OH_LABEL, INLINE_OH_LABEL = get_oh_labels(HOST)
async def is_v1_enabled_for_github_resolver(user_id: str) -> bool:
return await get_user_v1_enabled_setting(user_id) and ENABLE_V1_GITHUB_RESOLVER
async def get_user_proactive_conversation_setting(user_id: str | None) -> bool:
"""Get the user's proactive conversation setting.
@@ -68,56 +72,18 @@ async def get_user_proactive_conversation_setting(user_id: str | None) -> bool:
This function checks both the global environment variable kill switch AND
the user's individual setting. Both must be true for the function to return true.
"""
# If no user ID is provided, we can't check user settings
if not user_id:
return False
config = get_config()
settings_store = SaasSettingsStore(
user_id=user_id, session_maker=session_maker, config=config
)
settings = await call_sync_from_async(
settings_store.get_user_settings_by_keycloak_id, user_id
)
if not settings or settings.enable_proactive_conversation_starters is None:
# Check global setting first - if disabled globally, return False
if not ENABLE_PROACTIVE_CONVERSATION_STARTERS:
return False
return settings.enable_proactive_conversation_starters
async def get_user_v1_enabled_setting(user_id: str) -> bool:
"""Get the user's V1 conversation API setting.
Args:
user_id: The keycloak user ID
Returns:
True if V1 conversations are enabled for this user, False otherwise
Note:
This function checks both the global environment variable kill switch AND
the user's individual setting. Both must be true for the function to return true.
"""
# Check the global environment variable first
if not ENABLE_V1_GITHUB_RESOLVER:
org = await OrgStore.get_current_org_from_keycloak_user_id(user_id)
if not org:
return False
config = get_config()
settings_store = SaasSettingsStore(
user_id=user_id, session_maker=session_maker, config=config
)
settings = await call_sync_from_async(
settings_store.get_user_settings_by_keycloak_id, user_id
)
if not settings or settings.v1_enabled is None:
return False
return settings.v1_enabled
return bool(org.enable_proactive_conversation_starters)
# =================================================
@@ -140,7 +106,10 @@ class GithubIssue(ResolverViewInterface):
title: str
description: str
previous_comments: list[Comment]
v1: bool
v1_enabled: bool
def _get_branch_name(self) -> str | None:
return getattr(self, 'branch_name', None)
async def _load_resolver_context(self):
github_service = GithubServiceImpl(
@@ -175,12 +144,11 @@ class GithubIssue(ResolverViewInterface):
issue_body=self.description,
previous_comments=self.previous_comments,
)
return user_instructions, conversation_instructions
async def _get_user_secrets(self):
secrets_store = SaasSecretsStore(
self.user_info.keycloak_user_id, session_maker, get_config()
)
secrets_store = SaasSecretsStore(self.user_info.keycloak_user_id, get_config())
user_secrets = await secrets_store.load()
return user_secrets.custom_secrets if user_secrets else None
@@ -188,26 +156,32 @@ class GithubIssue(ResolverViewInterface):
async def initialize_new_conversation(self) -> ConversationMetadata:
# FIXME: Handle if initialize_conversation returns None
v1_enabled = await get_user_v1_enabled_setting(self.user_info.keycloak_user_id)
logger.info(
f'[GitHub V1]: User flag found for {self.user_info.keycloak_user_id} is {v1_enabled}'
self.v1_enabled = await is_v1_enabled_for_github_resolver(
self.user_info.keycloak_user_id
)
if v1_enabled:
logger.info(
f'[GitHub V1]: User flag found for {self.user_info.keycloak_user_id} is {self.v1_enabled}'
)
if self.v1_enabled:
# Create dummy conversationm metadata
# Don't save to conversation store
# V1 conversations are stored in a separate table
self.conversation_id = uuid4().hex
return ConversationMetadata(
conversation_id=uuid4().hex, selected_repository=self.full_repo_name
conversation_id=self.conversation_id,
selected_repository=self.full_repo_name,
)
conversation_metadata: ConversationMetadata = await initialize_conversation( # type: ignore[assignment]
user_id=self.user_info.keycloak_user_id,
conversation_id=None,
selected_repository=self.full_repo_name,
selected_branch=None,
selected_branch=self._get_branch_name(),
conversation_trigger=ConversationTrigger.RESOLVER,
git_provider=ProviderType.GITHUB,
)
self.conversation_id = conversation_metadata.conversation_id
return conversation_metadata
@@ -218,25 +192,18 @@ class GithubIssue(ResolverViewInterface):
conversation_metadata: ConversationMetadata,
saas_user_auth: UserAuth,
):
v1_enabled = await get_user_v1_enabled_setting(self.user_info.keycloak_user_id)
logger.info(
f'[GitHub V1]: User flag found for {self.user_info.keycloak_user_id} is {v1_enabled}'
)
if v1_enabled:
try:
# Use V1 app conversation service
await self._create_v1_conversation(
jinja_env, saas_user_auth, conversation_metadata
)
return
except Exception as e:
logger.warning(f'Error checking V1 settings, falling back to V0: {e}')
# Use existing V0 conversation service
await self._create_v0_conversation(
jinja_env, git_provider_tokens, conversation_metadata
f'[GitHub V1]: User flag found for {self.user_info.keycloak_user_id} is {self.v1_enabled}'
)
if self.v1_enabled:
# Use V1 app conversation service
await self._create_v1_conversation(
jinja_env, saas_user_auth, conversation_metadata
)
else:
await self._create_v0_conversation(
jinja_env, git_provider_tokens, conversation_metadata
)
async def _create_v0_conversation(
self,
@@ -264,6 +231,29 @@ class GithubIssue(ResolverViewInterface):
conversation_instructions=conversation_instructions,
)
async def _get_v1_initial_user_message(self, jinja_env: Environment) -> str:
"""Build the initial user message for V1 resolver conversations.
For "issue opened" events (no specific comment body), we can simply
concatenate the user prompt and the rendered issue context.
Subclasses that represent comment-driven events (issue comments, PR review
comments, inline review comments) override this method to control ordering
(e.g., context first, then the triggering comment, then previous comments).
"""
user_instructions, conversation_instructions = await self._get_instructions(
jinja_env
)
parts: list[str] = []
if user_instructions.strip():
parts.append(user_instructions.strip())
if conversation_instructions.strip():
parts.append(conversation_instructions.strip())
return '\n\n'.join(parts)
async def _create_v1_conversation(
self,
jinja_env: Environment,
@@ -273,13 +263,11 @@ class GithubIssue(ResolverViewInterface):
"""Create conversation using the new V1 app conversation system."""
logger.info('[GitHub V1]: Creating V1 conversation')
user_instructions, conversation_instructions = await self._get_instructions(
jinja_env
)
initial_user_text = await self._get_v1_initial_user_message(jinja_env)
# Create the initial message request
initial_message = SendMessageRequest(
role='user', content=[TextContent(text=user_instructions)]
role='user', content=[TextContent(text=initial_user_text)]
)
# Create the GitHub V1 callback processor
@@ -291,9 +279,12 @@ class GithubIssue(ResolverViewInterface):
# Create the V1 conversation start request with the callback processor
start_request = AppConversationStartRequest(
conversation_id=UUID(conversation_metadata.conversation_id),
system_message_suffix=conversation_instructions,
# NOTE: Resolver instructions are intended to be lower priority than the
# system prompt, so we inject them into the initial user message.
system_message_suffix=None,
initial_message=initial_message,
selected_repository=self.full_repo_name,
selected_branch=self._get_branch_name(),
git_provider=ProviderType.GITHUB,
title=f'GitHub Issue #{self.issue_number}: {self.title}',
trigger=ConversationTrigger.RESOLVER,
@@ -318,11 +309,9 @@ class GithubIssue(ResolverViewInterface):
f'Failed to start V1 conversation: {task.detail}'
)
self.v1 = True
def _create_github_v1_callback_processor(self):
"""Create a V1 callback processor for GitHub integration."""
from openhands.app_server.event_callback.github_v1_callback_processor import (
from integrations.github.github_v1_callback_processor import (
GithubV1CallbackProcessor,
)
@@ -354,7 +343,6 @@ class GithubIssueComment(GithubIssue):
conversation_instructions_template = jinja_env.get_template(
'issue_conversation_instructions.j2'
)
conversation_instructions = conversation_instructions_template.render(
issue_number=self.issue_number,
issue_title=self.title,
@@ -364,6 +352,17 @@ class GithubIssueComment(GithubIssue):
return user_instructions, conversation_instructions
async def _get_v1_initial_user_message(self, jinja_env: Environment) -> str:
await self._load_resolver_context()
template = jinja_env.get_template('issue_comment_initial_message.j2')
return template.render(
issue_number=self.issue_number,
issue_title=self.title,
issue_body=self.description,
issue_comment=self.comment_body,
previous_comments=self.previous_comments,
).strip()
@dataclass
class GithubPRComment(GithubIssueComment):
@@ -390,30 +389,17 @@ class GithubPRComment(GithubIssueComment):
return user_instructions, conversation_instructions
async def initialize_new_conversation(self) -> ConversationMetadata:
v1_enabled = await get_user_v1_enabled_setting(self.user_info.keycloak_user_id)
logger.info(
f'[GitHub V1]: User flag found for {self.user_info.keycloak_user_id} is {v1_enabled}'
)
if v1_enabled:
# Create dummy conversationm metadata
# Don't save to conversation store
# V1 conversations are stored in a separate table
return ConversationMetadata(
conversation_id=uuid4().hex, selected_repository=self.full_repo_name
)
conversation_metadata: ConversationMetadata = await initialize_conversation( # type: ignore[assignment]
user_id=self.user_info.keycloak_user_id,
conversation_id=None,
selected_repository=self.full_repo_name,
selected_branch=self.branch_name,
conversation_trigger=ConversationTrigger.RESOLVER,
git_provider=ProviderType.GITHUB,
)
self.conversation_id = conversation_metadata.conversation_id
return conversation_metadata
async def _get_v1_initial_user_message(self, jinja_env: Environment) -> str:
await self._load_resolver_context()
template = jinja_env.get_template('pr_update_initial_message.j2')
return template.render(
pr_number=self.issue_number,
branch_name=self.branch_name,
pr_title=self.title,
pr_body=self.description,
pr_comment=self.comment_body,
comments=self.previous_comments,
).strip()
@dataclass
@@ -449,7 +435,6 @@ class GithubInlinePRComment(GithubPRComment):
conversation_instructions_template = jinja_env.get_template(
'pr_update_conversation_instructions.j2'
)
conversation_instructions = conversation_instructions_template.render(
pr_number=self.issue_number,
pr_title=self.title,
@@ -462,9 +447,23 @@ class GithubInlinePRComment(GithubPRComment):
return user_instructions, conversation_instructions
async def _get_v1_initial_user_message(self, jinja_env: Environment) -> str:
await self._load_resolver_context()
template = jinja_env.get_template('pr_update_initial_message.j2')
return template.render(
pr_number=self.issue_number,
branch_name=self.branch_name,
pr_title=self.title,
pr_body=self.description,
file_location=self.file_location,
line_number=self.line_number,
pr_comment=self.comment_body,
comments=self.previous_comments,
).strip()
def _create_github_v1_callback_processor(self):
"""Create a V1 callback processor for GitHub integration."""
from openhands.app_server.event_callback.github_v1_callback_processor import (
from integrations.github.github_v1_callback_processor import (
GithubV1CallbackProcessor,
)
@@ -729,13 +728,13 @@ class GithubFactory:
def _interact_with_github() -> Issue | None:
with GithubIntegration(
GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
auth=Auth.AppAuth(GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY)
) as integration:
access_token = integration.get_access_token(
payload['installation']['id']
).token
with Github(access_token) as gh:
with Github(auth=Auth.Token(access_token)) as gh:
repo = gh.get_repo(selected_repo)
login = (
payload['organization']['login']
@@ -794,7 +793,7 @@ class GithubFactory:
@staticmethod
async def create_github_view_from_payload(
message: Message, keycloak_user_id: str
) -> ResolverViewInterface:
) -> GithubViewType:
"""Create the appropriate class (GithubIssue or GithubPRComment) based on the payload.
Also return metadata about the event (e.g., action type).
"""
@@ -830,7 +829,7 @@ class GithubFactory:
title='',
description='',
previous_comments=[],
v1=False,
v1_enabled=False,
)
elif GithubFactory.is_issue_comment(message):
@@ -856,7 +855,7 @@ class GithubFactory:
title='',
description='',
previous_comments=[],
v1=False,
v1_enabled=False,
)
elif GithubFactory.is_pr_comment(message):
@@ -867,12 +866,12 @@ class GithubFactory:
access_token = ''
with GithubIntegration(
GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY
auth=Auth.AppAuth(GITHUB_APP_CLIENT_ID, GITHUB_APP_PRIVATE_KEY)
) as integration:
access_token = integration.get_access_token(installation_id).token
head_ref = None
with Github(access_token) as gh:
with Github(auth=Auth.Token(access_token)) as gh:
repo = gh.get_repo(selected_repo)
pull_request = repo.get_pull(issue_number)
head_ref = pull_request.head.ref
@@ -898,7 +897,7 @@ class GithubFactory:
title='',
description='',
previous_comments=[],
v1=False,
v1_enabled=False,
)
elif GithubFactory.is_inline_pr_comment(message):
@@ -932,7 +931,7 @@ class GithubFactory:
title='',
description='',
previous_comments=[],
v1=False,
v1_enabled=False,
)
else:

View File

@@ -1,4 +1,7 @@
from __future__ import annotations
from types import MappingProxyType
from typing import cast
from integrations.gitlab.gitlab_view import (
GitlabFactory,
@@ -15,7 +18,9 @@ from integrations.utils import (
CONVERSATION_URL,
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
get_session_expired_message,
)
from integrations.v1_utils import get_saas_user_auth
from jinja2 import Environment, FileSystemLoader
from pydantic import SecretStr
from server.auth.token_manager import TokenManager
@@ -24,11 +29,15 @@ from server.utils.conversation_callback_utils import register_callback_processor
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.gitlab.gitlab_service import GitLabServiceImpl
from openhands.integrations.provider import ProviderToken, ProviderType
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.storage.data_models.secrets import Secrets
class GitlabManager(Manager):
class GitlabManager(Manager[GitlabViewType]):
def __init__(self, token_manager: TokenManager, data_collector: None = None):
self.token_manager = token_manager
@@ -62,11 +71,11 @@ class GitlabManager(Manager):
logger.warning(f'Got invalid keyloak user id for GitLab User {user_id}')
return False
# Importing here prevents circular import
# GitLabServiceImpl returns SaaSGitLabService in enterprise context
from integrations.gitlab.gitlab_service import SaaSGitLabService
gitlab_service: SaaSGitLabService = GitLabServiceImpl(
external_auth_id=keycloak_user_id
gitlab_service = cast(
SaaSGitLabService, GitLabServiceImpl(external_auth_id=keycloak_user_id)
)
return await gitlab_service.user_has_write_access(project_id)
@@ -116,55 +125,52 @@ class GitlabManager(Manager):
# Check if the user has write access to the repository
return has_write_access
async def send_message(self, message: Message, gitlab_view: ResolverViewInterface):
"""
Send a message to GitLab based on the view type.
async def send_message(self, message: str, gitlab_view: ResolverViewInterface):
"""Send a message to GitLab based on the view type.
Args:
message: The message to send
message: The message content to send (plain text string)
gitlab_view: The GitLab view object containing issue/PR/comment info
"""
keycloak_user_id = gitlab_view.user_info.keycloak_user_id
# Importing here prevents circular import
# GitLabServiceImpl returns SaaSGitLabService in enterprise context
from integrations.gitlab.gitlab_service import SaaSGitLabService
gitlab_service: SaaSGitLabService = GitLabServiceImpl(
external_auth_id=keycloak_user_id
gitlab_service = cast(
SaaSGitLabService, GitLabServiceImpl(external_auth_id=keycloak_user_id)
)
outgoing_message = message.message
if isinstance(gitlab_view, GitlabInlineMRComment) or isinstance(
gitlab_view, GitlabMRComment
):
await gitlab_service.reply_to_mr(
gitlab_view.project_id,
gitlab_view.issue_number,
gitlab_view.discussion_id,
message.message,
project_id=str(gitlab_view.project_id),
merge_request_iid=str(gitlab_view.issue_number),
discussion_id=gitlab_view.discussion_id,
body=message,
)
elif isinstance(gitlab_view, GitlabIssueComment):
await gitlab_service.reply_to_issue(
gitlab_view.project_id,
gitlab_view.issue_number,
gitlab_view.discussion_id,
outgoing_message,
project_id=str(gitlab_view.project_id),
issue_number=str(gitlab_view.issue_number),
discussion_id=gitlab_view.discussion_id,
body=message,
)
elif isinstance(gitlab_view, GitlabIssue):
await gitlab_service.reply_to_issue(
gitlab_view.project_id,
gitlab_view.issue_number,
None, # no discussion id, issue is tagged
outgoing_message,
project_id=str(gitlab_view.project_id),
issue_number=str(gitlab_view.issue_number),
discussion_id=None, # no discussion id, issue is tagged
body=message,
)
else:
logger.warning(
f'[GitLab] Unsupported view type: {type(gitlab_view).__name__}'
)
async def start_job(self, gitlab_view: GitlabViewType):
async def start_job(self, gitlab_view: GitlabViewType) -> None:
"""
Start a job for the GitLab view.
@@ -209,8 +215,18 @@ class GitlabManager(Manager):
)
)
# Initialize conversation and get metadata (following GitHub pattern)
convo_metadata = await gitlab_view.initialize_new_conversation()
saas_user_auth = await get_saas_user_auth(
gitlab_view.user_info.keycloak_user_id, self.token_manager
)
await gitlab_view.create_new_conversation(
self.jinja_env, secret_store.provider_tokens
self.jinja_env,
secret_store.provider_tokens,
convo_metadata,
saas_user_auth,
)
conversation_id = gitlab_view.conversation_id
@@ -219,18 +235,19 @@ class GitlabManager(Manager):
f'[GitLab] Created conversation {conversation_id} for user {user_info.username}'
)
# Create a GitlabCallbackProcessor for this conversation
processor = GitlabCallbackProcessor(
gitlab_view=gitlab_view,
send_summary_instruction=True,
)
if not gitlab_view.v1_enabled:
# Create a GitlabCallbackProcessor for this conversation
processor = GitlabCallbackProcessor(
gitlab_view=gitlab_view,
send_summary_instruction=True,
)
# Register the callback processor
register_callback_processor(conversation_id, processor)
# Register the callback processor
register_callback_processor(conversation_id, processor)
logger.info(
f'[GitLab] Created callback processor for conversation {conversation_id}'
)
logger.info(
f'[GitLab] Created callback processor for conversation {conversation_id}'
)
conversation_link = CONVERSATION_URL.format(conversation_id)
msg_info = f"I'm on it! {user_info.username} can [track my progress at all-hands.dev]({conversation_link})"
@@ -249,13 +266,18 @@ class GitlabManager(Manager):
msg_info = f'@{user_info.username} please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
except SessionExpiredError as e:
logger.warning(
f'[GitLab] Session expired for user {user_info.username}: {str(e)}'
)
msg_info = get_session_expired_message(user_info.username)
# Send the acknowledgment message
msg = self.create_outgoing_message(msg_info)
await self.send_message(msg, gitlab_view)
await self.send_message(msg_info, gitlab_view)
except Exception as e:
logger.exception(f'[GitLab] Error starting job: {str(e)}')
msg = self.create_outgoing_message(
msg='Uh oh! There was an unexpected error starting the job :('
await self.send_message(
'Uh oh! There was an unexpected error starting the job :(', gitlab_view
)
await self.send_message(msg, gitlab_view)

View File

@@ -1,7 +1,7 @@
import asyncio
from integrations.store_repo_utils import store_repositories_in_db
from integrations.types import GitLabResourceType
from integrations.utils import store_repositories_in_db
from pydantic import SecretStr
from server.auth.token_manager import TokenManager
from storage.gitlab_webhook import GitlabWebhook, WebhookStatus
@@ -80,22 +80,52 @@ class SaaSGitLabService(GitLabService):
logger.warning('external_auth_token and user_id not set!')
return gitlab_token
async def get_owned_groups(self) -> list[dict]:
async def get_owned_groups(self, min_access_level: int = 40) -> list[dict]:
"""
Get all groups for which the current user is the owner.
Get all top-level groups where the current user has admin access.
This method supports pagination and fetches all groups where the user has
at least the specified access level.
Args:
min_access_level: Minimum access level required (default: 40 for Maintainer or Owner)
- 40: Maintainer or Owner
- 50: Owner only
Returns:
list[dict]: A list of groups owned by the current user.
list[dict]: A list of groups where user has the specified access level or higher.
"""
url = f'{self.BASE_URL}/groups'
params = {'owned': 'true', 'per_page': 100, 'top_level_only': 'true'}
groups_with_admin_access = []
page = 1
per_page = 100
try:
response, headers = await self._make_request(url, params)
return response
except Exception:
logger.warning('Error fetching owned groups', exc_info=True)
return []
while True:
try:
url = f'{self.BASE_URL}/groups'
params = {
'page': str(page),
'per_page': str(per_page),
'min_access_level': min_access_level,
'top_level_only': 'true',
}
response, headers = await self._make_request(url, params)
if not response:
break
groups_with_admin_access.extend(response)
page += 1
# Check if we've reached the last page
link_header = headers.get('Link', '')
if 'rel="next"' not in link_header:
break
except Exception:
logger.warning(f'Error fetching groups on page {page}', exc_info=True)
break
return groups_with_admin_access
async def add_owned_projects_and_groups_to_db(self, owned_personal_projects):
"""
@@ -155,6 +185,30 @@ class SaaSGitLabService(GitLabService):
users_personal_projects: List of personal projects owned by the user
repositories: List of Repository objects to store
"""
# If external_auth_id is not set, try to determine it from the Keycloak token
if not self.external_auth_id and self.external_auth_token:
try:
user_info = await self.token_manager.get_user_info(
self.external_auth_token.get_secret_value()
)
keycloak_user_id = user_info.sub
self.external_auth_id = keycloak_user_id
logger.info(
f'Determined external_auth_id from Keycloak token: {self.external_auth_id}'
)
except Exception:
logger.warning(
'Cannot store repository data: external_auth_id is not set and could not be determined from token',
exc_info=True,
)
return
if not self.external_auth_id:
logger.warning(
'Cannot store repository data: external_auth_id could not be determined'
)
return
try:
# First, add owned projects and groups to the database
await self.add_owned_projects_and_groups_to_db(users_personal_projects)
@@ -527,3 +581,55 @@ class SaaSGitLabService(GitLabService):
await self._make_request(url=url, params=params, method=RequestMethod.POST)
except Exception as e:
logger.exception(f'[GitLab]: Reply to MR failed {e}')
async def get_user_resources_with_admin_access(
self,
) -> tuple[list[dict], list[dict]]:
"""
Get all projects and groups where the current user has admin access (maintainer or owner).
Returns:
tuple[list[dict], list[dict]]: A tuple containing:
- list of projects where user has admin access
- list of groups where user has admin access
"""
projects_with_admin_access = []
groups_with_admin_access = []
# Fetch all projects the user is a member of
page = 1
per_page = 100
while True:
try:
url = f'{self.BASE_URL}/projects'
params = {
'page': str(page),
'per_page': str(per_page),
'membership': 1,
'min_access_level': 40, # Maintainer or Owner
}
response, headers = await self._make_request(url, params)
if not response:
break
projects_with_admin_access.extend(response)
page += 1
# Check if we've reached the last page
link_header = headers.get('Link', '')
if 'rel="next"' not in link_header:
break
except Exception:
logger.warning(f'Error fetching projects on page {page}', exc_info=True)
break
# Fetch all groups where user is owner or maintainer
groups_with_admin_access = await self.get_owned_groups(min_access_level=40)
logger.info(
f'Found {len(projects_with_admin_access)} projects and {len(groups_with_admin_access)} groups with admin access'
)
return projects_with_admin_access, groups_with_admin_access

View File

@@ -0,0 +1,269 @@
import logging
from typing import Any
from uuid import UUID
import httpx
from integrations.utils import get_summary_instruction
from integrations.v1_utils import handle_callback_error
from pydantic import Field
from openhands.agent_server.models import AskAgentRequest, AskAgentResponse
from openhands.app_server.event_callback.event_callback_models import (
EventCallback,
EventCallbackProcessor,
)
from openhands.app_server.event_callback.event_callback_result_models import (
EventCallbackResult,
EventCallbackResultStatus,
)
from openhands.app_server.event_callback.util import (
ensure_conversation_found,
ensure_running_sandbox,
get_agent_server_url_from_sandbox,
)
from openhands.sdk import Event
from openhands.sdk.event import ConversationStateUpdateEvent
_logger = logging.getLogger(__name__)
class GitlabV1CallbackProcessor(EventCallbackProcessor):
"""Callback processor for GitLab V1 integrations."""
gitlab_view_data: dict[str, Any] = Field(default_factory=dict)
should_request_summary: bool = Field(default=True)
inline_mr_comment: bool = Field(default=False)
async def __call__(
self,
conversation_id: UUID,
callback: EventCallback,
event: Event,
) -> EventCallbackResult | None:
"""Process events for GitLab V1 integration."""
# Only handle ConversationStateUpdateEvent
if not isinstance(event, ConversationStateUpdateEvent):
return None
# Only act when execution has finished
if not (event.key == 'execution_status' and event.value == 'finished'):
return None
_logger.info('[GitLab V1] Callback agent state was %s', event)
_logger.info(
'[GitLab V1] Should request summary: %s', self.should_request_summary
)
if not self.should_request_summary:
return None
self.should_request_summary = False
try:
_logger.info(f'[GitLab V1] Requesting summary {conversation_id}')
summary = await self._request_summary(conversation_id)
_logger.info(
f'[GitLab V1] Posting summary {conversation_id}',
extra={'summary': summary},
)
await self._post_summary_to_gitlab(summary)
return EventCallbackResult(
status=EventCallbackResultStatus.SUCCESS,
event_callback_id=callback.id,
event_id=event.id,
conversation_id=conversation_id,
detail=summary,
)
except Exception as e:
can_post_error = bool(self.gitlab_view_data.get('keycloak_user_id'))
await handle_callback_error(
error=e,
conversation_id=conversation_id,
service_name='GitLab',
service_logger=_logger,
can_post_error=can_post_error,
post_error_func=self._post_summary_to_gitlab,
)
return EventCallbackResult(
status=EventCallbackResultStatus.ERROR,
event_callback_id=callback.id,
event_id=event.id,
conversation_id=conversation_id,
detail=str(e),
)
# -------------------------------------------------------------------------
# GitLab helpers
# -------------------------------------------------------------------------
async def _post_summary_to_gitlab(self, summary: str) -> None:
"""Post a summary comment to the configured GitLab issue or MR."""
# Import here to avoid circular imports
from integrations.gitlab.gitlab_service import SaaSGitLabService
keycloak_user_id = self.gitlab_view_data.get('keycloak_user_id')
if not keycloak_user_id:
raise RuntimeError('Missing keycloak user ID for GitLab')
gitlab_service = SaaSGitLabService(external_auth_id=keycloak_user_id)
project_id = self.gitlab_view_data['project_id']
issue_number = self.gitlab_view_data['issue_number']
discussion_id = self.gitlab_view_data['discussion_id']
is_mr = self.gitlab_view_data.get('is_mr', False)
if is_mr:
await gitlab_service.reply_to_mr(
project_id,
issue_number,
discussion_id,
summary,
)
else:
await gitlab_service.reply_to_issue(
project_id,
issue_number,
discussion_id,
summary,
)
# -------------------------------------------------------------------------
# Agent / sandbox helpers
# -------------------------------------------------------------------------
async def _ask_question(
self,
httpx_client: httpx.AsyncClient,
agent_server_url: str,
conversation_id: UUID,
session_api_key: str,
message_content: str,
) -> str:
"""Send a message to the agent server via the V1 API and return response text."""
send_message_request = AskAgentRequest(question=message_content)
url = (
f"{agent_server_url.rstrip('/')}"
f"/api/conversations/{conversation_id}/ask_agent"
)
headers = {'X-Session-API-Key': session_api_key}
payload = send_message_request.model_dump()
try:
response = await httpx_client.post(
url,
json=payload,
headers=headers,
timeout=30.0,
)
response.raise_for_status()
agent_response = AskAgentResponse.model_validate(response.json())
return agent_response.response
except httpx.HTTPStatusError as e:
error_detail = f'HTTP {e.response.status_code} error'
try:
error_body = e.response.text
if error_body:
error_detail += f': {error_body}'
except Exception: # noqa: BLE001
pass
_logger.error(
'[GitLab V1] HTTP error sending message to %s: %s. '
'Request payload: %s. Response headers: %s',
url,
error_detail,
payload,
dict(e.response.headers),
exc_info=True,
)
raise Exception(f'Failed to send message to agent server: {error_detail}')
except httpx.TimeoutException:
error_detail = f'Request timeout after 30 seconds to {url}'
_logger.error(
'[GitLab V1] %s. Request payload: %s',
error_detail,
payload,
exc_info=True,
)
raise Exception(error_detail)
except httpx.RequestError as e:
error_detail = f'Request error to {url}: {str(e)}'
_logger.error(
'[GitLab V1] %s. Request payload: %s',
error_detail,
payload,
exc_info=True,
)
raise Exception(error_detail)
# -------------------------------------------------------------------------
# Summary orchestration
# -------------------------------------------------------------------------
async def _request_summary(self, conversation_id: UUID) -> str:
"""Ask the agent to produce a summary of its work and return the agent response.
NOTE: This method now returns a string (the agent server's response text)
and raises exceptions on errors. The wrapping into EventCallbackResult
is handled by __call__.
"""
# Import services within the method to avoid circular imports
from openhands.app_server.config import (
get_app_conversation_info_service,
get_httpx_client,
get_sandbox_service,
)
from openhands.app_server.services.injector import InjectorState
from openhands.app_server.user.specifiy_user_context import (
ADMIN,
USER_CONTEXT_ATTR,
)
# Create injector state for dependency injection
state = InjectorState()
setattr(state, USER_CONTEXT_ATTR, ADMIN)
async with (
get_app_conversation_info_service(state) as app_conversation_info_service,
get_sandbox_service(state) as sandbox_service,
get_httpx_client(state) as httpx_client,
):
# 1. Conversation lookup
app_conversation_info = ensure_conversation_found(
await app_conversation_info_service.get_app_conversation_info(
conversation_id
),
conversation_id,
)
# 2. Sandbox lookup + validation
sandbox = ensure_running_sandbox(
await sandbox_service.get_sandbox(app_conversation_info.sandbox_id),
app_conversation_info.sandbox_id,
)
assert (
sandbox.session_api_key is not None
), f'No session API key for sandbox: {sandbox.id}'
# 3. URL + instruction
agent_server_url = get_agent_server_url_from_sandbox(sandbox)
# Prepare message based on agent state
message_content = get_summary_instruction()
# Ask the agent and return the response text
return await self._ask_question(
httpx_client=httpx_client,
agent_server_url=agent_server_url,
conversation_id=conversation_id,
session_api_key=sandbox.session_api_key,
message_content=message_content,
)

View File

@@ -1,25 +1,53 @@
from dataclasses import dataclass
from uuid import UUID, uuid4
from integrations.models import Message
from integrations.resolver_context import ResolverUserContext
from integrations.types import ResolverViewInterface, UserData
from integrations.utils import HOST, get_oh_labels, has_exact_mention
from integrations.utils import (
ENABLE_V1_GITLAB_RESOLVER,
HOST,
get_oh_labels,
get_user_v1_enabled_setting,
has_exact_mention,
)
from jinja2 import Environment
from server.auth.token_manager import TokenManager
from server.config import get_config
from storage.database import session_maker
from storage.saas_secrets_store import SaasSecretsStore
from openhands.agent_server.models import SendMessageRequest
from openhands.app_server.app_conversation.app_conversation_models import (
AppConversationStartRequest,
AppConversationStartTaskStatus,
)
from openhands.app_server.config import get_app_conversation_service
from openhands.app_server.services.injector import InjectorState
from openhands.app_server.user.specifiy_user_context import USER_CONTEXT_ATTR
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.gitlab.gitlab_service import GitLabServiceImpl
from openhands.integrations.provider import PROVIDER_TOKEN_TYPE, ProviderType
from openhands.integrations.service_types import Comment
from openhands.server.services.conversation_service import create_new_conversation
from openhands.storage.data_models.conversation_metadata import ConversationTrigger
from openhands.sdk import TextContent
from openhands.server.services.conversation_service import (
initialize_conversation,
start_conversation,
)
from openhands.server.user_auth.user_auth import UserAuth
from openhands.storage.data_models.conversation_metadata import (
ConversationMetadata,
ConversationTrigger,
)
OH_LABEL, INLINE_OH_LABEL = get_oh_labels(HOST)
CONFIDENTIAL_NOTE = 'confidential_note'
NOTE_TYPES = ['note', CONFIDENTIAL_NOTE]
async def is_v1_enabled_for_gitlab_resolver(user_id: str) -> bool:
return await get_user_v1_enabled_setting(user_id) and ENABLE_V1_GITLAB_RESOLVER
# =================================================
# SECTION: Factory to create appriorate Gitlab view
# =================================================
@@ -41,6 +69,10 @@ class GitlabIssue(ResolverViewInterface):
description: str
previous_comments: list[Comment]
is_mr: bool
v1_enabled: bool
def _get_branch_name(self) -> str | None:
return getattr(self, 'branch_name', None)
async def _load_resolver_context(self):
gitlab_service = GitLabServiceImpl(
@@ -78,35 +110,158 @@ class GitlabIssue(ResolverViewInterface):
return user_instructions, conversation_instructions
async def _get_user_secrets(self):
secrets_store = SaasSecretsStore(
self.user_info.keycloak_user_id, session_maker, get_config()
)
secrets_store = SaasSecretsStore(self.user_info.keycloak_user_id, get_config())
user_secrets = await secrets_store.load()
return user_secrets.custom_secrets if user_secrets else None
async def initialize_new_conversation(self) -> ConversationMetadata:
# v1_enabled is already set at construction time in the factory method
# This is the source of truth for the conversation type
if self.v1_enabled:
# Create dummy conversation metadata
# Don't save to conversation store
# V1 conversations are stored in a separate table
self.conversation_id = uuid4().hex
return ConversationMetadata(
conversation_id=self.conversation_id,
selected_repository=self.full_repo_name,
)
conversation_metadata: ConversationMetadata = await initialize_conversation( # type: ignore[assignment]
user_id=self.user_info.keycloak_user_id,
conversation_id=None,
selected_repository=self.full_repo_name,
selected_branch=self._get_branch_name(),
conversation_trigger=ConversationTrigger.RESOLVER,
git_provider=ProviderType.GITLAB,
)
self.conversation_id = conversation_metadata.conversation_id
return conversation_metadata
async def create_new_conversation(
self, jinja_env: Environment, git_provider_tokens: PROVIDER_TOKEN_TYPE
self,
jinja_env: Environment,
git_provider_tokens: PROVIDER_TOKEN_TYPE,
conversation_metadata: ConversationMetadata,
saas_user_auth: UserAuth,
):
# v1_enabled is already set at construction time in the factory method
if self.v1_enabled:
# Use V1 app conversation service
await self._create_v1_conversation(
jinja_env, saas_user_auth, conversation_metadata
)
else:
await self._create_v0_conversation(
jinja_env, git_provider_tokens, conversation_metadata
)
async def _create_v0_conversation(
self,
jinja_env: Environment,
git_provider_tokens: PROVIDER_TOKEN_TYPE,
conversation_metadata: ConversationMetadata,
):
"""Create conversation using the legacy V0 system."""
logger.info('[GitLab]: Creating V0 conversation')
custom_secrets = await self._get_user_secrets()
user_instructions, conversation_instructions = await self._get_instructions(
jinja_env
)
agent_loop_info = await create_new_conversation(
await start_conversation(
user_id=self.user_info.keycloak_user_id,
git_provider_tokens=git_provider_tokens,
custom_secrets=custom_secrets,
selected_repository=self.full_repo_name,
selected_branch=None,
initial_user_msg=user_instructions,
conversation_instructions=conversation_instructions,
image_urls=None,
conversation_trigger=ConversationTrigger.RESOLVER,
replay_json=None,
conversation_id=conversation_metadata.conversation_id,
conversation_metadata=conversation_metadata,
conversation_instructions=conversation_instructions,
)
async def _create_v1_conversation(
self,
jinja_env: Environment,
saas_user_auth: UserAuth,
conversation_metadata: ConversationMetadata,
):
"""Create conversation using the new V1 app conversation system."""
logger.info('[GitLab V1]: Creating V1 conversation')
user_instructions, conversation_instructions = await self._get_instructions(
jinja_env
)
# Create the initial message request
initial_message = SendMessageRequest(
role='user', content=[TextContent(text=user_instructions)]
)
# Create the GitLab V1 callback processor
gitlab_callback_processor = self._create_gitlab_v1_callback_processor()
# Get the app conversation service and start the conversation
injector_state = InjectorState()
# Determine the title based on whether it's an MR or issue
title_prefix = 'GitLab MR' if self.is_mr else 'GitLab Issue'
title = f'{title_prefix} #{self.issue_number}: {self.title}'
# Create the V1 conversation start request with the callback processor
start_request = AppConversationStartRequest(
conversation_id=UUID(conversation_metadata.conversation_id),
system_message_suffix=conversation_instructions,
initial_message=initial_message,
selected_repository=self.full_repo_name,
selected_branch=self._get_branch_name(),
git_provider=ProviderType.GITLAB,
title=title,
trigger=ConversationTrigger.RESOLVER,
processors=[
gitlab_callback_processor
], # Pass the callback processor directly
)
# Set up the GitLab user context for the V1 system
gitlab_user_context = ResolverUserContext(saas_user_auth=saas_user_auth)
setattr(injector_state, USER_CONTEXT_ATTR, gitlab_user_context)
async with get_app_conversation_service(
injector_state
) as app_conversation_service:
async for task in app_conversation_service.start_app_conversation(
start_request
):
if task.status == AppConversationStartTaskStatus.ERROR:
logger.error(f'Failed to start V1 conversation: {task.detail}')
raise RuntimeError(
f'Failed to start V1 conversation: {task.detail}'
)
def _create_gitlab_v1_callback_processor(self):
"""Create a V1 callback processor for GitLab integration."""
from integrations.gitlab.gitlab_v1_callback_processor import (
GitlabV1CallbackProcessor,
)
# Create and return the GitLab V1 callback processor
return GitlabV1CallbackProcessor(
gitlab_view_data={
'issue_number': self.issue_number,
'project_id': self.project_id,
'full_repo_name': self.full_repo_name,
'installation_id': self.installation_id,
'keycloak_user_id': self.user_info.keycloak_user_id,
'is_mr': self.is_mr,
'discussion_id': getattr(self, 'discussion_id', None),
},
send_summary_instruction=self.send_summary_instruction,
)
self.conversation_id = agent_loop_info.conversation_id
return self.conversation_id
@dataclass
@@ -141,6 +296,9 @@ class GitlabIssueComment(GitlabIssue):
class GitlabMRComment(GitlabIssueComment):
branch_name: str
def _get_branch_name(self) -> str | None:
return self.branch_name
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
user_instructions_template = jinja_env.get_template('mr_update_prompt.j2')
await self._load_resolver_context()
@@ -162,29 +320,6 @@ class GitlabMRComment(GitlabIssueComment):
return user_instructions, conversation_instructions
async def create_new_conversation(
self, jinja_env: Environment, git_provider_tokens: PROVIDER_TOKEN_TYPE
):
custom_secrets = await self._get_user_secrets()
user_instructions, conversation_instructions = await self._get_instructions(
jinja_env
)
agent_loop_info = await create_new_conversation(
user_id=self.user_info.keycloak_user_id,
git_provider_tokens=git_provider_tokens,
custom_secrets=custom_secrets,
selected_repository=self.full_repo_name,
selected_branch=self.branch_name,
initial_user_msg=user_instructions,
conversation_instructions=conversation_instructions,
image_urls=None,
conversation_trigger=ConversationTrigger.RESOLVER,
replay_json=None,
)
self.conversation_id = agent_loop_info.conversation_id
return self.conversation_id
@dataclass
class GitlabInlineMRComment(GitlabMRComment):
@@ -306,7 +441,7 @@ class GitlabFactory:
@staticmethod
async def create_gitlab_view_from_payload(
message: Message, token_manager: TokenManager
) -> ResolverViewInterface:
) -> GitlabViewType:
payload = message.message['payload']
installation_id = message.message['installation_id']
user = payload['user']
@@ -325,6 +460,16 @@ class GitlabFactory:
user_id=user_id, username=username, keycloak_user_id=keycloak_user_id
)
# Check v1_enabled at construction time - this is the source of truth
v1_enabled = (
await is_v1_enabled_for_gitlab_resolver(keycloak_user_id)
if keycloak_user_id
else False
)
logger.info(
f'[GitLab V1]: User flag found for {keycloak_user_id} is {v1_enabled}'
)
if GitlabFactory.is_labeled_issue(message):
issue_iid = payload['object_attributes']['iid']
@@ -346,6 +491,7 @@ class GitlabFactory:
description='',
previous_comments=[],
is_mr=False,
v1_enabled=v1_enabled,
)
elif GitlabFactory.is_issue_comment(message):
@@ -376,6 +522,7 @@ class GitlabFactory:
description='',
previous_comments=[],
is_mr=False,
v1_enabled=v1_enabled,
)
elif GitlabFactory.is_mr_comment(message):
@@ -408,6 +555,7 @@ class GitlabFactory:
description='',
previous_comments=[],
is_mr=True,
v1_enabled=v1_enabled,
)
elif GitlabFactory.is_mr_comment(message, inline=True):
@@ -448,4 +596,7 @@ class GitlabFactory:
description='',
previous_comments=[],
is_mr=True,
v1_enabled=v1_enabled,
)
raise ValueError(f'Unhandled GitLab webhook event: {message}')

View File

@@ -0,0 +1,194 @@
"""Shared utilities for GitLab webhook installation.
This module contains reusable functions and classes for installing GitLab webhooks
that can be used by both the cron job and API routes.
"""
from __future__ import annotations
from typing import TYPE_CHECKING
from uuid import uuid4
from integrations.types import GitLabResourceType
from integrations.utils import GITLAB_WEBHOOK_URL
from storage.gitlab_webhook import GitlabWebhook, WebhookStatus
from storage.gitlab_webhook_store import GitlabWebhookStore
from openhands.core.logger import openhands_logger as logger
if TYPE_CHECKING:
from integrations.gitlab.gitlab_service import SaaSGitLabService
# Webhook configuration constants
WEBHOOK_NAME = 'OpenHands Resolver'
SCOPES: list[str] = [
'note_events',
'merge_requests_events',
'confidential_issues_events',
'issues_events',
'confidential_note_events',
'job_events',
'pipeline_events',
]
class BreakLoopException(Exception):
"""Exception raised when webhook installation conditions are not met or rate limited."""
pass
async def verify_webhook_conditions(
gitlab_service: SaaSGitLabService,
resource_type: GitLabResourceType,
resource_id: str,
webhook_store: GitlabWebhookStore,
webhook: GitlabWebhook,
) -> None:
"""
Verify all conditions are met for webhook installation.
Raises BreakLoopException if any condition fails or rate limited.
Args:
gitlab_service: GitLab service instance
resource_type: Type of resource (PROJECT or GROUP)
resource_id: ID of the resource
webhook_store: Webhook store instance
webhook: Webhook object to verify
"""
# Check if resource exists
does_resource_exist, status = await gitlab_service.check_resource_exists(
resource_type, resource_id
)
logger.info(
'Does resource exists',
extra={
'does_resource_exist': does_resource_exist,
'status': status,
'resource_id': resource_id,
'resource_type': resource_type,
},
)
if status == WebhookStatus.RATE_LIMITED:
raise BreakLoopException()
if not does_resource_exist and status != WebhookStatus.RATE_LIMITED:
await webhook_store.delete_webhook(webhook)
raise BreakLoopException()
# Check if user has admin access
(
is_user_admin_of_resource,
status,
) = await gitlab_service.check_user_has_admin_access_to_resource(
resource_type, resource_id
)
logger.info(
'Is user admin',
extra={
'is_user_admin': is_user_admin_of_resource,
'status': status,
'resource_id': resource_id,
'resource_type': resource_type,
},
)
if status == WebhookStatus.RATE_LIMITED:
raise BreakLoopException()
if not is_user_admin_of_resource:
await webhook_store.delete_webhook(webhook)
raise BreakLoopException()
# Check if webhook already exists
(
does_webhook_exist_on_resource,
status,
) = await gitlab_service.check_webhook_exists_on_resource(
resource_type=resource_type,
resource_id=resource_id,
webhook_url=GITLAB_WEBHOOK_URL,
)
logger.info(
'Does webhook already exist',
extra={
'does_webhook_exist_on_resource': does_webhook_exist_on_resource,
'status': status,
'resource_id': resource_id,
'resource_type': resource_type,
},
)
if status == WebhookStatus.RATE_LIMITED:
raise BreakLoopException()
if does_webhook_exist_on_resource != webhook.webhook_exists:
await webhook_store.update_webhook(
webhook, {'webhook_exists': does_webhook_exist_on_resource}
)
if does_webhook_exist_on_resource:
raise BreakLoopException()
async def install_webhook_on_resource(
gitlab_service: SaaSGitLabService,
resource_type: GitLabResourceType,
resource_id: str,
webhook_store: GitlabWebhookStore,
webhook: GitlabWebhook,
) -> tuple[str | None, WebhookStatus | None]:
"""
Install webhook on a GitLab resource.
Args:
gitlab_service: GitLab service instance
resource_type: Type of resource (PROJECT or GROUP)
resource_id: ID of the resource
webhook_store: Webhook store instance
webhook: Webhook object to install
Returns:
Tuple of (webhook_id, status)
"""
webhook_secret = f'{webhook.user_id}-{str(uuid4())}'
webhook_uuid = f'{str(uuid4())}'
webhook_id, status = await gitlab_service.install_webhook(
resource_type=resource_type,
resource_id=resource_id,
webhook_name=WEBHOOK_NAME,
webhook_url=GITLAB_WEBHOOK_URL,
webhook_secret=webhook_secret,
webhook_uuid=webhook_uuid,
scopes=SCOPES,
)
log_extra = {
'webhook_id': webhook_id,
'status': status,
'resource_id': resource_id,
'resource_type': resource_type,
}
if status == WebhookStatus.RATE_LIMITED:
logger.warning('Rate limited while creating webhook', extra=log_extra)
raise BreakLoopException()
if webhook_id:
await webhook_store.update_webhook(
webhook=webhook,
update_fields={
'webhook_secret': webhook_secret,
'webhook_exists': True, # webhook was created
'webhook_url': GITLAB_WEBHOOK_URL,
'scopes': SCOPES,
'webhook_uuid': webhook_uuid, # required to identify which webhook installation is sending payload
},
)
logger.info('Created new webhook', extra=log_extra)
else:
logger.error('Failed to create webhook', extra=log_extra)
return webhook_id, status

View File

@@ -1,22 +1,38 @@
import hashlib
import hmac
from typing import Dict, Optional, Tuple
from urllib.parse import urlparse
"""Jira integration manager.
This module orchestrates the processing of Jira webhook events:
1. Parse webhook payload (via JiraPayloadParser)
2. Validate workspace
3. Authenticate user
4. Create view with repository selection (via JiraFactory)
5. Start conversation job
The manager delegates payload parsing to JiraPayloadParser and view creation
to JiraFactory, keeping the orchestration logic clean and traceable.
"""
import httpx
from fastapi import Request
from integrations.jira.jira_types import JiraViewInterface
from integrations.jira.jira_view import (
JiraExistingConversationView,
JiraFactory,
JiraNewConversationView,
from integrations.jira.jira_payload import (
JiraPayloadError,
JiraPayloadParser,
JiraPayloadSkipped,
JiraPayloadSuccess,
JiraWebhookPayload,
)
from integrations.jira.jira_types import (
JiraViewInterface,
RepositoryNotFoundError,
StartingConvoException,
)
from integrations.jira.jira_view import JiraFactory, JiraNewConversationView
from integrations.manager import Manager
from integrations.models import JobContext, Message
from integrations.models import Message
from integrations.utils import (
HOST,
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
filter_potential_repos_by_user_msg,
get_oh_labels,
get_session_expired_message,
)
from jinja2 import Environment, FileSystemLoader
from server.auth.saas_user_auth import get_user_auth_from_keycloak_id
@@ -27,312 +43,221 @@ from storage.jira_user import JiraUser
from storage.jira_workspace import JiraWorkspace
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.provider import ProviderHandler
from openhands.integrations.service_types import Repository
from openhands.server.shared import server_config
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.server.user_auth.user_auth import UserAuth
from openhands.utils.http_session import httpx_verify_option
JIRA_CLOUD_API_URL = 'https://api.atlassian.com/ex/jira'
# Get OH labels for this environment
OH_LABEL, INLINE_OH_LABEL = get_oh_labels(HOST)
class JiraManager(Manager[JiraViewInterface]):
"""Manager for processing Jira webhook events.
This class orchestrates the flow from webhook receipt to conversation creation,
delegating parsing to JiraPayloadParser and view creation to JiraFactory.
"""
class JiraManager(Manager):
def __init__(self, token_manager: TokenManager):
self.token_manager = token_manager
self.integration_store = JiraIntegrationStore.get_instance()
self.jinja_env = Environment(
loader=FileSystemLoader(OPENHANDS_RESOLVER_TEMPLATES_DIR + 'jira')
)
self.payload_parser = JiraPayloadParser(
oh_label=OH_LABEL,
inline_oh_label=INLINE_OH_LABEL,
)
async def authenticate_user(
self, jira_user_id: str, workspace_id: int
async def receive_message(self, message: Message):
"""Process incoming Jira webhook message.
Flow:
1. Parse webhook payload
2. Validate workspace exists and is active
3. Authenticate user
4. Create view (includes fetching issue details and selecting repository)
5. Start job
Each step has clear logging for traceability.
"""
raw_payload = message.message.get('payload', {})
# Step 1: Parse webhook payload
logger.info(
'[Jira] Received webhook',
extra={'raw_payload': raw_payload},
)
parse_result = self.payload_parser.parse(raw_payload)
if isinstance(parse_result, JiraPayloadSkipped):
logger.info(
'[Jira] Webhook skipped', extra={'reason': parse_result.skip_reason}
)
return
if isinstance(parse_result, JiraPayloadError):
logger.warning(
'[Jira] Webhook parse failed', extra={'error': parse_result.error}
)
return
payload = parse_result.payload
logger.info(
'[Jira] Processing webhook',
extra={
'event_type': payload.event_type.value,
'issue_key': payload.issue_key,
'user_email': payload.user_email,
},
)
# Step 2: Validate workspace
workspace = await self._get_active_workspace(payload)
if not workspace:
return
# Step 3: Authenticate user
jira_user, saas_user_auth = await self._authenticate_user(payload, workspace)
if not jira_user or not saas_user_auth:
return
# Step 4: Create view (includes issue details fetch and repo selection)
decrypted_api_key = self.token_manager.decrypt_text(workspace.svc_acc_api_key)
try:
view = await JiraFactory.create_view(
payload=payload,
workspace=workspace,
user=jira_user,
user_auth=saas_user_auth,
decrypted_api_key=decrypted_api_key,
)
except RepositoryNotFoundError as e:
logger.warning(
'[Jira] Repository not found',
extra={'issue_key': payload.issue_key, 'error': str(e)},
)
await self._send_error_from_payload(payload, workspace, str(e))
return
except StartingConvoException as e:
logger.warning(
'[Jira] View creation failed',
extra={'issue_key': payload.issue_key, 'error': str(e)},
)
await self._send_error_from_payload(payload, workspace, str(e))
return
except Exception as e:
logger.error(
'[Jira] Unexpected error creating view',
extra={'issue_key': payload.issue_key, 'error': str(e)},
exc_info=True,
)
await self._send_error_from_payload(
payload,
workspace,
'Failed to initialize conversation. Please try again.',
)
return
# Step 5: Start job
await self.start_job(view)
async def _get_active_workspace(
self, payload: JiraWebhookPayload
) -> JiraWorkspace | None:
"""Validate and return the workspace for the webhook.
Returns None if:
- Workspace not found
- Workspace is inactive
- Request is from service account (to prevent recursion)
"""
workspace = await self.integration_store.get_workspace_by_name(
payload.workspace_name
)
if not workspace:
logger.warning(
'[Jira] Workspace not found',
extra={'workspace_name': payload.workspace_name},
)
# Can't send error without workspace credentials
return None
# Prevent recursive triggers from service account
if payload.user_email == workspace.svc_acc_email:
logger.debug(
'[Jira] Ignoring service account trigger',
extra={'workspace_name': payload.workspace_name},
)
return None
if workspace.status != 'active':
logger.warning(
'[Jira] Workspace inactive',
extra={'workspace_id': workspace.id, 'status': workspace.status},
)
await self._send_error_from_payload(
payload, workspace, 'Jira integration is not active for your workspace.'
)
return None
return workspace
async def _authenticate_user(
self, payload: JiraWebhookPayload, workspace: JiraWorkspace
) -> tuple[JiraUser | None, UserAuth | None]:
"""Authenticate Jira user and get their OpenHands user auth."""
# Find active Jira user by Keycloak user ID and workspace ID
"""Authenticate the Jira user and get OpenHands auth."""
jira_user = await self.integration_store.get_active_user(
jira_user_id, workspace_id
payload.account_id, workspace.id
)
if not jira_user:
logger.warning(
f'[Jira] No active Jira user found for {jira_user_id} in workspace {workspace_id}'
'[Jira] User not found or inactive',
extra={
'account_id': payload.account_id,
'user_email': payload.user_email,
'workspace_id': workspace.id,
},
)
await self._send_error_from_payload(
payload,
workspace,
f'User {payload.user_email} is not authenticated or active in the Jira integration.',
)
return None, None
saas_user_auth = await get_user_auth_from_keycloak_id(
jira_user.keycloak_user_id
)
if not saas_user_auth:
logger.warning(
'[Jira] Failed to get OpenHands auth',
extra={
'keycloak_user_id': jira_user.keycloak_user_id,
'user_email': payload.user_email,
},
)
await self._send_error_from_payload(
payload,
workspace,
f'User {payload.user_email} is not authenticated with OpenHands.',
)
return None, None
return jira_user, saas_user_auth
async def _get_repositories(self, user_auth: UserAuth) -> list[Repository]:
"""Get repositories that the user has access to."""
provider_tokens = await user_auth.get_provider_tokens()
if provider_tokens is None:
return []
access_token = await user_auth.get_access_token()
user_id = await user_auth.get_user_id()
client = ProviderHandler(
provider_tokens=provider_tokens,
external_auth_token=access_token,
external_auth_id=user_id,
)
repos: list[Repository] = await client.get_repositories(
'pushed', server_config.app_mode, None, None, None, None
)
return repos
async def validate_request(
self, request: Request
) -> Tuple[bool, Optional[str], Optional[Dict]]:
"""Verify Jira webhook signature."""
signature_header = request.headers.get('x-hub-signature')
signature = signature_header.split('=')[1] if signature_header else None
body = await request.body()
payload = await request.json()
workspace_name = ''
if payload.get('webhookEvent') == 'comment_created':
selfUrl = payload.get('comment', {}).get('author', {}).get('self')
elif payload.get('webhookEvent') == 'jira:issue_updated':
selfUrl = payload.get('user', {}).get('self')
else:
workspace_name = ''
parsedUrl = urlparse(selfUrl)
if parsedUrl.hostname:
workspace_name = parsedUrl.hostname
if not workspace_name:
logger.warning('[Jira] No workspace name found in webhook payload')
return False, None, None
if not signature:
logger.warning('[Jira] No signature found in webhook headers')
return False, None, None
workspace = await self.integration_store.get_workspace_by_name(workspace_name)
if not workspace:
logger.warning('[Jira] Could not identify workspace for webhook')
return False, None, None
if workspace.status != 'active':
logger.warning(f'[Jira] Workspace {workspace.id} is not active')
return False, None, None
webhook_secret = self.token_manager.decrypt_text(workspace.webhook_secret)
digest = hmac.new(webhook_secret.encode(), body, hashlib.sha256).hexdigest()
if hmac.compare_digest(signature, digest):
logger.info('[Jira] Webhook signature verified successfully')
return True, signature, payload
return False, None, None
def parse_webhook(self, payload: Dict) -> JobContext | None:
event_type = payload.get('webhookEvent')
if event_type == 'comment_created':
comment_data = payload.get('comment', {})
comment = comment_data.get('body', '')
if '@openhands' not in comment:
return None
issue_data = payload.get('issue', {})
issue_id = issue_data.get('id')
issue_key = issue_data.get('key')
base_api_url = issue_data.get('self', '').split('/rest/')[0]
user_data = comment_data.get('author', {})
user_email = user_data.get('emailAddress')
display_name = user_data.get('displayName')
account_id = user_data.get('accountId')
elif event_type == 'jira:issue_updated':
changelog = payload.get('changelog', {})
items = changelog.get('items', [])
labels = [
item.get('toString', '')
for item in items
if item.get('field') == 'labels' and 'toString' in item
]
if 'openhands' not in labels:
return None
issue_data = payload.get('issue', {})
issue_id = issue_data.get('id')
issue_key = issue_data.get('key')
base_api_url = issue_data.get('self', '').split('/rest/')[0]
user_data = payload.get('user', {})
user_email = user_data.get('emailAddress')
display_name = user_data.get('displayName')
account_id = user_data.get('accountId')
comment = ''
else:
return None
workspace_name = ''
parsedUrl = urlparse(base_api_url)
if parsedUrl.hostname:
workspace_name = parsedUrl.hostname
if not all(
[
issue_id,
issue_key,
user_email,
display_name,
account_id,
workspace_name,
base_api_url,
]
):
return None
return JobContext(
issue_id=issue_id,
issue_key=issue_key,
user_msg=comment,
user_email=user_email,
display_name=display_name,
platform_user_id=account_id,
workspace_name=workspace_name,
base_api_url=base_api_url,
)
async def receive_message(self, message: Message):
"""Process incoming Jira webhook message."""
payload = message.message.get('payload', {})
job_context = self.parse_webhook(payload)
if not job_context:
logger.info('[Jira] Webhook does not match trigger conditions')
return
# Get workspace by user email domain
workspace = await self.integration_store.get_workspace_by_name(
job_context.workspace_name
)
if not workspace:
logger.warning(
f'[Jira] No workspace found for email domain: {job_context.user_email}'
)
await self._send_error_comment(
job_context,
'Your workspace is not configured with Jira integration.',
None,
)
return
# Prevent any recursive triggers from the service account
if job_context.user_email == workspace.svc_acc_email:
return
if workspace.status != 'active':
logger.warning(f'[Jira] Workspace {workspace.id} is not active')
await self._send_error_comment(
job_context,
'Jira integration is not active for your workspace.',
workspace,
)
return
# Authenticate user
jira_user, saas_user_auth = await self.authenticate_user(
job_context.platform_user_id, workspace.id
)
if not jira_user or not saas_user_auth:
logger.warning(
f'[Jira] User authentication failed for {job_context.user_email}'
)
await self._send_error_comment(
job_context,
f'User {job_context.user_email} is not authenticated or active in the Jira integration.',
workspace,
)
return
# Get issue details
try:
api_key = self.token_manager.decrypt_text(workspace.svc_acc_api_key)
issue_title, issue_description = await self.get_issue_details(
job_context, workspace.jira_cloud_id, workspace.svc_acc_email, api_key
)
job_context.issue_title = issue_title
job_context.issue_description = issue_description
except Exception as e:
logger.error(f'[Jira] Failed to get issue context: {str(e)}')
await self._send_error_comment(
job_context,
'Failed to retrieve issue details. Please check the issue key and try again.',
workspace,
)
return
try:
# Create Jira view
jira_view = await JiraFactory.create_jira_view_from_payload(
job_context,
saas_user_auth,
jira_user,
workspace,
)
except Exception as e:
logger.error(f'[Jira] Failed to create jira view: {str(e)}', exc_info=True)
await self._send_error_comment(
job_context,
'Failed to initialize conversation. Please try again.',
workspace,
)
return
if not await self.is_job_requested(message, jira_view):
return
await self.start_job(jira_view)
async def is_job_requested(
self, message: Message, jira_view: JiraViewInterface
) -> bool:
"""
Check if a job is requested and handle repository selection.
"""
if isinstance(jira_view, JiraExistingConversationView):
return True
try:
# Get user repositories
user_repos: list[Repository] = await self._get_repositories(
jira_view.saas_user_auth
)
target_str = f'{jira_view.job_context.issue_description}\n{jira_view.job_context.user_msg}'
# Try to infer repository from issue description
match, repos = filter_potential_repos_by_user_msg(target_str, user_repos)
if match:
# Found exact repository match
jira_view.selected_repo = repos[0].full_name
logger.info(f'[Jira] Inferred repository: {repos[0].full_name}')
return True
else:
# No clear match - send repository selection comment
await self._send_repo_selection_comment(jira_view)
return False
except Exception as e:
logger.error(f'[Jira] Error in is_job_requested: {str(e)}')
return False
async def start_job(self, jira_view: JiraViewInterface):
async def start_job(self, view: JiraViewInterface) -> None:
"""Start a Jira job/conversation."""
# Import here to prevent circular import
from server.conversation_callback_processor.jira_callback_processor import (
@@ -340,110 +265,101 @@ class JiraManager(Manager):
)
try:
user_info: JiraUser = jira_view.jira_user
logger.info(
f'[Jira] Starting job for user {user_info.keycloak_user_id} '
f'issue {jira_view.job_context.issue_key}',
'[Jira] Starting job',
extra={
'issue_key': view.payload.issue_key,
'user_id': view.jira_user.keycloak_user_id,
'selected_repo': view.selected_repo,
},
)
# Create conversation
conversation_id = await jira_view.create_or_update_conversation(
self.jinja_env
)
conversation_id = await view.create_or_update_conversation(self.jinja_env)
logger.info(
f'[Jira] Created/Updated conversation {conversation_id} for issue {jira_view.job_context.issue_key}'
'[Jira] Conversation created',
extra={
'conversation_id': conversation_id,
'issue_key': view.payload.issue_key,
},
)
# Register callback processor for updates
if isinstance(jira_view, JiraNewConversationView):
if isinstance(view, JiraNewConversationView):
processor = JiraCallbackProcessor(
issue_key=jira_view.job_context.issue_key,
workspace_name=jira_view.jira_workspace.name,
issue_key=view.payload.issue_key,
workspace_name=view.jira_workspace.name,
)
# Register the callback processor
register_callback_processor(conversation_id, processor)
logger.info(
f'[Jira] Created callback processor for conversation {conversation_id}'
'[Jira] Callback processor registered',
extra={'conversation_id': conversation_id},
)
# Send initial response
msg_info = jira_view.get_response_msg()
# Send success response
msg_info = view.get_response_msg()
except MissingSettingsError as e:
logger.warning(f'[Jira] Missing settings error: {str(e)}')
logger.warning(
'[Jira] Missing settings error',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
)
msg_info = f'Please re-login into [OpenHands Cloud]({HOST_URL}) before starting a job.'
except LLMAuthenticationError as e:
logger.warning(f'[Jira] LLM authentication error: {str(e)}')
logger.warning(
'[Jira] LLM authentication error',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
)
msg_info = f'Please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
except SessionExpiredError as e:
logger.warning(
'[Jira] Session expired',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
)
msg_info = get_session_expired_message()
except StartingConvoException as e:
logger.warning(
'[Jira] Conversation start failed',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
)
msg_info = str(e)
except Exception as e:
logger.error(
f'[Jira] Unexpected error starting job: {str(e)}', exc_info=True
'[Jira] Unexpected error starting job',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
exc_info=True,
)
msg_info = 'Sorry, there was an unexpected error starting the job. Please try again.'
# Send response comment
try:
api_key = self.token_manager.decrypt_text(
jira_view.jira_workspace.svc_acc_api_key
)
await self.send_message(
self.create_outgoing_message(msg=msg_info),
issue_key=jira_view.job_context.issue_key,
jira_cloud_id=jira_view.jira_workspace.jira_cloud_id,
svc_acc_email=jira_view.jira_workspace.svc_acc_email,
svc_acc_api_key=api_key,
)
except Exception as e:
logger.error(f'[Jira] Failed to send response message: {str(e)}')
async def get_issue_details(
self,
job_context: JobContext,
jira_cloud_id: str,
svc_acc_email: str,
svc_acc_api_key: str,
) -> Tuple[str, str]:
url = f'{JIRA_CLOUD_API_URL}/{jira_cloud_id}/rest/api/2/issue/{job_context.issue_key}'
async with httpx.AsyncClient(verify=httpx_verify_option()) as client:
response = await client.get(url, auth=(svc_acc_email, svc_acc_api_key))
response.raise_for_status()
issue_payload = response.json()
if not issue_payload:
raise ValueError(f'Issue with key {job_context.issue_key} not found.')
title = issue_payload.get('fields', {}).get('summary', '')
description = issue_payload.get('fields', {}).get('description', '')
if not title:
raise ValueError(
f'Issue with key {job_context.issue_key} does not have a title.'
)
if not description:
raise ValueError(
f'Issue with key {job_context.issue_key} does not have a description.'
)
return title, description
await self._send_comment(view, msg_info)
async def send_message(
self,
message: Message,
message: str,
issue_key: str,
jira_cloud_id: str,
svc_acc_email: str,
svc_acc_api_key: str,
):
"""Send a comment to a Jira issue.
Args:
message: The message content to send (plain text string)
issue_key: The Jira issue key (e.g., 'PROJ-123')
jira_cloud_id: The Jira Cloud ID
svc_acc_email: Service account email for authentication
svc_acc_api_key: Service account API key for authentication
"""
url = (
f'{JIRA_CLOUD_API_URL}/{jira_cloud_id}/rest/api/2/issue/{issue_key}/comment'
)
data = {'body': message.message}
data = {'body': message}
async with httpx.AsyncClient(verify=httpx_verify_option()) as client:
response = await client.post(
url, auth=(svc_acc_email, svc_acc_api_key), json=data
@@ -451,54 +367,53 @@ class JiraManager(Manager):
response.raise_for_status()
return response.json()
async def _send_error_comment(
self,
job_context: JobContext,
error_msg: str,
workspace: JiraWorkspace | None,
):
"""Send error comment to Jira issue."""
if not workspace:
logger.error('[Jira] Cannot send error comment - no workspace available')
return
async def _send_comment(self, view: JiraViewInterface, msg: str):
"""Send a comment using credentials from the view."""
try:
api_key = self.token_manager.decrypt_text(
view.jira_workspace.svc_acc_api_key
)
await self.send_message(
msg,
issue_key=view.payload.issue_key,
jira_cloud_id=view.jira_workspace.jira_cloud_id,
svc_acc_email=view.jira_workspace.svc_acc_email,
svc_acc_api_key=api_key,
)
except Exception as e:
logger.error(
'[Jira] Failed to send comment',
extra={'issue_key': view.payload.issue_key, 'error': str(e)},
)
async def _send_error_from_payload(
self,
payload: JiraWebhookPayload,
workspace: JiraWorkspace,
error_msg: str,
):
"""Send error comment before view is created (using payload directly)."""
try:
api_key = self.token_manager.decrypt_text(workspace.svc_acc_api_key)
await self.send_message(
self.create_outgoing_message(msg=error_msg),
issue_key=job_context.issue_key,
error_msg,
issue_key=payload.issue_key,
jira_cloud_id=workspace.jira_cloud_id,
svc_acc_email=workspace.svc_acc_email,
svc_acc_api_key=api_key,
)
except Exception as e:
logger.error(f'[Jira] Failed to send error comment: {str(e)}')
async def _send_repo_selection_comment(self, jira_view: JiraViewInterface):
"""Send a comment with repository options for the user to choose."""
try:
comment_msg = (
'I need to know which repository to work with. '
'Please add it to your issue description or send a followup comment.'
)
api_key = self.token_manager.decrypt_text(
jira_view.jira_workspace.svc_acc_api_key
)
await self.send_message(
self.create_outgoing_message(msg=comment_msg),
issue_key=jira_view.job_context.issue_key,
jira_cloud_id=jira_view.jira_workspace.jira_cloud_id,
svc_acc_email=jira_view.jira_workspace.svc_acc_email,
svc_acc_api_key=api_key,
)
logger.info(
f'[Jira] Sent repository selection comment for issue {jira_view.job_context.issue_key}'
)
except Exception as e:
logger.error(
f'[Jira] Failed to send repository selection comment: {str(e)}'
'[Jira] Failed to send error comment',
extra={'issue_key': payload.issue_key, 'error': str(e)},
)
def get_workspace_name_from_payload(self, payload: dict) -> str | None:
"""Extract workspace name from Jira webhook payload.
This method is used by the route for signature verification.
"""
parse_result = self.payload_parser.parse(payload)
if isinstance(parse_result, JiraPayloadSuccess):
return parse_result.payload.workspace_name
return None

View File

@@ -0,0 +1,265 @@
"""Centralized payload parsing for Jira webhooks.
This module provides a single source of truth for parsing and validating
Jira webhook payloads, replacing scattered parsing logic throughout the codebase.
"""
from dataclasses import dataclass
from enum import Enum
from urllib.parse import urlparse
from openhands.core.logger import openhands_logger as logger
class JiraEventType(Enum):
"""Types of Jira events we handle."""
LABELED_TICKET = 'labeled_ticket'
COMMENT_MENTION = 'comment_mention'
@dataclass(frozen=True)
class JiraWebhookPayload:
"""Normalized, validated representation of a Jira webhook payload.
This immutable dataclass replaces JobContext and provides a single
source of truth for all webhook data. All parsing happens in
JiraPayloadParser, ensuring consistent validation.
"""
event_type: JiraEventType
raw_event: str # Original webhookEvent value
# Issue data
issue_id: str
issue_key: str
# User data
user_email: str
display_name: str
account_id: str
# Workspace data (derived from issue self URL)
workspace_name: str
base_api_url: str
# Event-specific data
comment_body: str = '' # For comment events
@property
def user_msg(self) -> str:
"""Alias for comment_body for backward compatibility."""
return self.comment_body
class JiraPayloadParseError(Exception):
"""Raised when payload parsing fails."""
def __init__(self, reason: str, event_type: str | None = None):
self.reason = reason
self.event_type = event_type
super().__init__(reason)
@dataclass(frozen=True)
class JiraPayloadSuccess:
"""Result when parsing succeeds."""
payload: JiraWebhookPayload
@dataclass(frozen=True)
class JiraPayloadSkipped:
"""Result when event is intentionally skipped."""
skip_reason: str
@dataclass(frozen=True)
class JiraPayloadError:
"""Result when parsing fails due to invalid data."""
error: str
JiraPayloadParseResult = JiraPayloadSuccess | JiraPayloadSkipped | JiraPayloadError
class JiraPayloadParser:
"""Centralized parser for Jira webhook payloads.
This class provides a single entry point for parsing webhooks,
determining event types, and extracting all necessary fields.
Replaces scattered parsing in JiraFactory and JiraManager.
"""
def __init__(self, oh_label: str, inline_oh_label: str):
"""Initialize parser with OpenHands label configuration.
Args:
oh_label: Label that triggers OpenHands (e.g., 'openhands')
inline_oh_label: Mention that triggers OpenHands (e.g., '@openhands')
"""
self.oh_label = oh_label
self.inline_oh_label = inline_oh_label
def parse(self, raw_payload: dict) -> JiraPayloadParseResult:
"""Parse a raw webhook payload into a normalized JiraWebhookPayload.
Args:
raw_payload: The raw webhook payload dict from Jira
Returns:
One of:
- JiraPayloadSuccess: Valid, actionable event with payload
- JiraPayloadSkipped: Event we intentionally don't process
- JiraPayloadError: Malformed payload we expected to process
"""
webhook_event = raw_payload.get('webhookEvent', '')
logger.debug(
'[Jira] Parsing webhook payload', extra={'webhook_event': webhook_event}
)
if webhook_event == 'jira:issue_updated':
return self._parse_label_event(raw_payload, webhook_event)
elif webhook_event == 'comment_created':
return self._parse_comment_event(raw_payload, webhook_event)
else:
return JiraPayloadSkipped(f'Unhandled webhook event type: {webhook_event}')
def _parse_label_event(
self, payload: dict, webhook_event: str
) -> JiraPayloadParseResult:
"""Parse an issue_updated event for label changes."""
changelog = payload.get('changelog', {})
items = changelog.get('items', [])
# Extract labels that were added
labels = [
item.get('toString', '')
for item in items
if item.get('field') == 'labels' and 'toString' in item
]
if self.oh_label not in labels:
return JiraPayloadSkipped(
f"Label event does not contain '{self.oh_label}' label"
)
# For label events, user data comes from 'user' field
user_data = payload.get('user', {})
return self._extract_and_validate(
payload=payload,
user_data=user_data,
event_type=JiraEventType.LABELED_TICKET,
webhook_event=webhook_event,
comment_body='',
)
def _parse_comment_event(
self, payload: dict, webhook_event: str
) -> JiraPayloadParseResult:
"""Parse a comment_created event."""
comment_data = payload.get('comment', {})
comment_body = comment_data.get('body', '')
if not self._has_mention(comment_body):
return JiraPayloadSkipped(
f"Comment does not mention '{self.inline_oh_label}'"
)
# For comment events, user data comes from 'comment.author'
user_data = comment_data.get('author', {})
return self._extract_and_validate(
payload=payload,
user_data=user_data,
event_type=JiraEventType.COMMENT_MENTION,
webhook_event=webhook_event,
comment_body=comment_body,
)
def _has_mention(self, text: str) -> bool:
"""Check if text contains an exact mention of OpenHands."""
from integrations.utils import has_exact_mention
return has_exact_mention(text, self.inline_oh_label)
def _extract_and_validate(
self,
payload: dict,
user_data: dict,
event_type: JiraEventType,
webhook_event: str,
comment_body: str,
) -> JiraPayloadParseResult:
"""Extract common fields and validate required data is present."""
issue_data = payload.get('issue', {})
# Extract all fields with empty string defaults (makes them str type)
issue_id = issue_data.get('id', '')
issue_key = issue_data.get('key', '')
user_email = user_data.get('emailAddress', '')
display_name = user_data.get('displayName', '')
account_id = user_data.get('accountId', '')
base_api_url, workspace_name = self._extract_workspace_from_url(
issue_data.get('self', '')
)
# Validate required fields
missing: list[str] = []
if not issue_id:
missing.append('issue.id')
if not issue_key:
missing.append('issue.key')
if not display_name:
missing.append('user.displayName')
if not account_id:
missing.append('user.accountId')
if not workspace_name:
missing.append('workspace_name (derived from issue.self)')
if not base_api_url:
missing.append('base_api_url (derived from issue.self)')
if missing:
return JiraPayloadError(f"Missing required fields: {', '.join(missing)}")
return JiraPayloadSuccess(
JiraWebhookPayload(
event_type=event_type,
raw_event=webhook_event,
issue_id=issue_id,
issue_key=issue_key,
user_email=user_email,
display_name=display_name,
account_id=account_id,
workspace_name=workspace_name,
base_api_url=base_api_url,
comment_body=comment_body,
)
)
def _extract_workspace_from_url(self, self_url: str) -> tuple[str, str]:
"""Extract base API URL and workspace name from issue self URL.
Args:
self_url: The 'self' URL from the issue data
Returns:
Tuple of (base_api_url, workspace_name)
"""
if not self_url:
return '', ''
# Extract base URL (everything before /rest/)
if '/rest/' in self_url:
base_api_url = self_url.split('/rest/')[0]
else:
parsed = urlparse(self_url)
base_api_url = f'{parsed.scheme}://{parsed.netloc}'
# Extract workspace name (hostname)
parsed = urlparse(base_api_url)
workspace_name = parsed.hostname or ''
return base_api_url, workspace_name

View File

@@ -1,26 +1,42 @@
from abc import ABC, abstractmethod
"""Type definitions and interfaces for Jira integration."""
from abc import ABC, abstractmethod
from typing import TYPE_CHECKING
from integrations.models import JobContext
from jinja2 import Environment
from storage.jira_user import JiraUser
from storage.jira_workspace import JiraWorkspace
from openhands.server.user_auth.user_auth import UserAuth
if TYPE_CHECKING:
from integrations.jira.jira_payload import JiraWebhookPayload
class JiraViewInterface(ABC):
"""Interface for Jira views that handle different types of Jira interactions."""
"""Interface for Jira views that handle different types of Jira interactions.
job_context: JobContext
Views hold the webhook payload directly rather than duplicating fields,
and fetch issue details lazily when needed.
"""
# Core data - view holds these references
payload: 'JiraWebhookPayload'
saas_user_auth: UserAuth
jira_user: JiraUser
jira_workspace: JiraWorkspace
# Mutable state set during processing
selected_repo: str | None
conversation_id: str
@abstractmethod
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Get initial instructions for the conversation."""
async def get_issue_details(self) -> tuple[str, str]:
"""Fetch and cache issue title and description from Jira API.
Returns:
Tuple of (issue_title, issue_description)
"""
pass
@abstractmethod
@@ -35,6 +51,21 @@ class JiraViewInterface(ABC):
class StartingConvoException(Exception):
"""Exception raised when starting a conversation fails."""
"""Exception raised when starting a conversation fails.
This provides user-friendly error messages that can be sent back to Jira.
"""
pass
class RepositoryNotFoundError(Exception):
"""Raised when a repository cannot be determined from the issue.
This is a separate error domain from StartingConvoException - it represents
a precondition failure (no repo configured/found) rather than a conversation
creation failure. The manager catches this and converts it to a user-friendly
message.
"""
pass

View File

@@ -1,8 +1,21 @@
from dataclasses import dataclass
"""Jira view implementations and factory.
from integrations.jira.jira_types import JiraViewInterface, StartingConvoException
from integrations.models import JobContext
from integrations.utils import CONVERSATION_URL, get_final_agent_observation
Views are responsible for:
- Holding the webhook payload and auth context
- Lazy-loading issue details from Jira API when needed
- Creating conversations with the selected repository
"""
from dataclasses import dataclass, field
import httpx
from integrations.jira.jira_payload import JiraWebhookPayload
from integrations.jira.jira_types import (
JiraViewInterface,
RepositoryNotFoundError,
StartingConvoException,
)
from integrations.utils import CONVERSATION_URL, infer_repo_from_message
from jinja2 import Environment
from storage.jira_conversation import JiraConversation
from storage.jira_integration_store import JiraIntegrationStore
@@ -10,55 +23,147 @@ from storage.jira_user import JiraUser
from storage.jira_workspace import JiraWorkspace
from openhands.core.logger import openhands_logger as logger
from openhands.core.schema.agent import AgentState
from openhands.events.action import MessageAction
from openhands.events.serialization.event import event_to_dict
from openhands.server.services.conversation_service import (
create_new_conversation,
setup_init_conversation_settings,
)
from openhands.server.shared import ConversationStoreImpl, config, conversation_manager
from openhands.integrations.provider import ProviderHandler
from openhands.server.services.conversation_service import create_new_conversation
from openhands.server.user_auth.user_auth import UserAuth
from openhands.storage.data_models.conversation_metadata import ConversationTrigger
from openhands.utils.http_session import httpx_verify_option
JIRA_CLOUD_API_URL = 'https://api.atlassian.com/ex/jira'
integration_store = JiraIntegrationStore.get_instance()
@dataclass
class JiraNewConversationView(JiraViewInterface):
job_context: JobContext
"""View for creating a new Jira conversation.
This view holds the webhook payload directly and lazily fetches
issue details when needed for rendering templates.
"""
payload: JiraWebhookPayload
saas_user_auth: UserAuth
jira_user: JiraUser
jira_workspace: JiraWorkspace
selected_repo: str | None
conversation_id: str
selected_repo: str | None = None
conversation_id: str = ''
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
# Lazy-loaded issue details (cached after first fetch)
_issue_title: str | None = field(default=None, repr=False)
_issue_description: str | None = field(default=None, repr=False)
# Decrypted API key (set by factory)
_decrypted_api_key: str = field(default='', repr=False)
async def get_issue_details(self) -> tuple[str, str]:
"""Fetch issue details from Jira API (cached after first call).
Returns:
Tuple of (issue_title, issue_description)
Raises:
StartingConvoException: If issue details cannot be fetched
"""
if self._issue_title is not None and self._issue_description is not None:
return self._issue_title, self._issue_description
try:
url = f'{JIRA_CLOUD_API_URL}/{self.jira_workspace.jira_cloud_id}/rest/api/2/issue/{self.payload.issue_key}'
async with httpx.AsyncClient(verify=httpx_verify_option()) as client:
response = await client.get(
url,
auth=(
self.jira_workspace.svc_acc_email,
self._decrypted_api_key,
),
)
response.raise_for_status()
issue_payload = response.json()
if not issue_payload:
raise StartingConvoException(
f'Issue {self.payload.issue_key} not found.'
)
self._issue_title = issue_payload.get('fields', {}).get('summary', '')
self._issue_description = (
issue_payload.get('fields', {}).get('description', '') or ''
)
if not self._issue_title:
raise StartingConvoException(
f'Issue {self.payload.issue_key} does not have a title.'
)
logger.info(
'[Jira] Fetched issue details',
extra={
'issue_key': self.payload.issue_key,
'has_description': bool(self._issue_description),
},
)
return self._issue_title, self._issue_description
except httpx.HTTPStatusError as e:
logger.error(
'[Jira] Failed to fetch issue details',
extra={
'issue_key': self.payload.issue_key,
'status': e.response.status_code,
},
)
raise StartingConvoException(
f'Failed to fetch issue details: HTTP {e.response.status_code}'
)
except Exception as e:
if isinstance(e, StartingConvoException):
raise
logger.error(
'[Jira] Failed to fetch issue details',
extra={'issue_key': self.payload.issue_key, 'error': str(e)},
)
raise StartingConvoException(f'Failed to fetch issue details: {str(e)}')
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Get instructions for the conversation.
This fetches issue details if not already cached.
Returns:
Tuple of (system_instructions, user_message)
"""
issue_title, issue_description = await self.get_issue_details()
instructions_template = jinja_env.get_template('jira_instructions.j2')
instructions = instructions_template.render()
user_msg_template = jinja_env.get_template('jira_new_conversation.j2')
user_msg = user_msg_template.render(
issue_key=self.job_context.issue_key,
issue_title=self.job_context.issue_title,
issue_description=self.job_context.issue_description,
user_message=self.job_context.user_msg or '',
issue_key=self.payload.issue_key,
issue_title=issue_title,
issue_description=issue_description,
user_message=self.payload.user_msg,
)
return instructions, user_msg
async def create_or_update_conversation(self, jinja_env: Environment) -> str:
"""Create a new Jira conversation"""
"""Create a new Jira conversation.
Returns:
The conversation ID
Raises:
StartingConvoException: If conversation creation fails
"""
if not self.selected_repo:
raise StartingConvoException('No repository selected for this conversation')
provider_tokens = await self.saas_user_auth.get_provider_tokens()
user_secrets = await self.saas_user_auth.get_secrets()
instructions, user_msg = self._get_instructions(jinja_env)
instructions, user_msg = await self._get_instructions(jinja_env)
try:
agent_loop_info = await create_new_conversation(
@@ -76,149 +181,259 @@ class JiraNewConversationView(JiraViewInterface):
self.conversation_id = agent_loop_info.conversation_id
logger.info(f'[Jira] Created conversation {self.conversation_id}')
logger.info(
'[Jira] Created conversation',
extra={
'conversation_id': self.conversation_id,
'issue_key': self.payload.issue_key,
'selected_repo': self.selected_repo,
},
)
# Store Jira conversation mapping
jira_conversation = JiraConversation(
conversation_id=self.conversation_id,
issue_id=self.job_context.issue_id,
issue_key=self.job_context.issue_key,
issue_id=self.payload.issue_id,
issue_key=self.payload.issue_key,
jira_user_id=self.jira_user.id,
)
await integration_store.create_conversation(jira_conversation)
return self.conversation_id
except Exception as e:
if isinstance(e, StartingConvoException):
raise
logger.error(
f'[Jira] Failed to create conversation: {str(e)}', exc_info=True
'[Jira] Failed to create conversation',
extra={'issue_key': self.payload.issue_key, 'error': str(e)},
exc_info=True,
)
raise StartingConvoException(f'Failed to create conversation: {str(e)}')
def get_response_msg(self) -> str:
"""Get the response message to send back to Jira"""
"""Get the response message to send back to Jira."""
conversation_link = CONVERSATION_URL.format(self.conversation_id)
return f"I'm on it! {self.job_context.display_name} can [track my progress here|{conversation_link}]."
@dataclass
class JiraExistingConversationView(JiraViewInterface):
job_context: JobContext
saas_user_auth: UserAuth
jira_user: JiraUser
jira_workspace: JiraWorkspace
selected_repo: str | None
conversation_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
user_msg_template = jinja_env.get_template('jira_existing_conversation.j2')
user_msg = user_msg_template.render(
issue_key=self.job_context.issue_key,
user_message=self.job_context.user_msg or '',
issue_title=self.job_context.issue_title,
issue_description=self.job_context.issue_description,
)
return '', user_msg
async def create_or_update_conversation(self, jinja_env: Environment) -> str:
"""Update an existing Jira conversation"""
user_id = self.jira_user.keycloak_user_id
try:
conversation_store = await ConversationStoreImpl.get_instance(
config, user_id
)
try:
await conversation_store.get_metadata(self.conversation_id)
except FileNotFoundError:
raise StartingConvoException('Conversation no longer exists.')
provider_tokens = await self.saas_user_auth.get_provider_tokens()
# Should we raise here if there are no providers?
providers_set = list(provider_tokens.keys()) if provider_tokens else []
conversation_init_data = await setup_init_conversation_settings(
user_id, self.conversation_id, providers_set
)
# Either join ongoing conversation, or restart the conversation
agent_loop_info = await conversation_manager.maybe_start_agent_loop(
self.conversation_id, conversation_init_data, user_id
)
final_agent_observation = get_final_agent_observation(
agent_loop_info.event_store
)
agent_state = (
None
if len(final_agent_observation) == 0
else final_agent_observation[0].agent_state
)
if not agent_state or agent_state == AgentState.LOADING:
raise StartingConvoException('Conversation is still starting')
_, user_msg = self._get_instructions(jinja_env)
user_message_event = MessageAction(content=user_msg)
await conversation_manager.send_event_to_conversation(
self.conversation_id, event_to_dict(user_message_event)
)
return self.conversation_id
except Exception as e:
logger.error(
f'[Jira] Failed to create conversation: {str(e)}', exc_info=True
)
raise StartingConvoException(f'Failed to create conversation: {str(e)}')
def get_response_msg(self) -> str:
"""Get the response message to send back to Jira"""
conversation_link = CONVERSATION_URL.format(self.conversation_id)
return f"I'm on it! {self.job_context.display_name} can [continue tracking my progress here|{conversation_link}]."
return f"I'm on it! {self.payload.display_name} can [track my progress here|{conversation_link}]."
class JiraFactory:
"""Factory for creating Jira views based on message content"""
"""Factory for creating Jira views.
The factory is responsible for:
- Creating the appropriate view type
- Inferring and selecting the repository
- Validating all required data is available
Repository selection happens here so that view creation either
succeeds with a valid repo or fails with a clear error.
"""
@staticmethod
async def create_jira_view_from_payload(
job_context: JobContext,
saas_user_auth: UserAuth,
jira_user: JiraUser,
jira_workspace: JiraWorkspace,
async def _create_provider_handler(user_auth: UserAuth) -> ProviderHandler | None:
"""Create a ProviderHandler for the user."""
provider_tokens = await user_auth.get_provider_tokens()
if provider_tokens is None:
return None
access_token = await user_auth.get_access_token()
user_id = await user_auth.get_user_id()
return ProviderHandler(
provider_tokens=provider_tokens,
external_auth_token=access_token,
external_auth_id=user_id,
)
@staticmethod
def _extract_potential_repos(
issue_key: str,
issue_title: str,
issue_description: str,
user_msg: str,
) -> list[str]:
"""Extract potential repository names from issue content.
Raises:
RepositoryNotFoundError: If no potential repos found in text.
"""
search_text = f'{issue_title}\n{issue_description}\n{user_msg}'
potential_repos = infer_repo_from_message(search_text)
if not potential_repos:
raise RepositoryNotFoundError(
'Could not determine which repository to use. '
'Please mention the repository (e.g., owner/repo) in the issue description or comment.'
)
logger.info(
'[Jira] Found potential repositories in issue content',
extra={'issue_key': issue_key, 'potential_repos': potential_repos},
)
return potential_repos
@staticmethod
async def _verify_repos(
issue_key: str,
potential_repos: list[str],
provider_handler: ProviderHandler,
) -> list[str]:
"""Verify which repos the user has access to."""
verified_repos: list[str] = []
for repo_name in potential_repos:
try:
repository = await provider_handler.verify_repo_provider(repo_name)
verified_repos.append(repository.full_name)
logger.debug(
'[Jira] Repository verification succeeded',
extra={'issue_key': issue_key, 'repository': repository.full_name},
)
except Exception as e:
logger.debug(
'[Jira] Repository verification failed',
extra={
'issue_key': issue_key,
'repo_name': repo_name,
'error': str(e),
},
)
return verified_repos
@staticmethod
def _select_single_repo(
issue_key: str,
potential_repos: list[str],
verified_repos: list[str],
) -> str:
"""Select exactly one repo from verified repos.
Raises:
RepositoryNotFoundError: If zero or multiple repos verified.
"""
if len(verified_repos) == 0:
raise RepositoryNotFoundError(
f'Could not access any of the mentioned repositories: {", ".join(potential_repos)}. '
'Please ensure you have access to the repository and it exists.'
)
if len(verified_repos) > 1:
raise RepositoryNotFoundError(
f'Multiple repositories found: {", ".join(verified_repos)}. '
'Please specify exactly one repository in the issue description or comment.'
)
logger.info(
'[Jira] Verified repository access',
extra={'issue_key': issue_key, 'repository': verified_repos[0]},
)
return verified_repos[0]
@staticmethod
async def _infer_repository(
payload: JiraWebhookPayload,
user_auth: UserAuth,
issue_title: str,
issue_description: str,
) -> str:
"""Infer and verify the repository from issue content.
Raises:
RepositoryNotFoundError: If no valid repository can be determined.
"""
provider_handler = await JiraFactory._create_provider_handler(user_auth)
if not provider_handler:
raise RepositoryNotFoundError(
'No Git provider connected. Please connect a Git provider in OpenHands settings.'
)
potential_repos = JiraFactory._extract_potential_repos(
payload.issue_key, issue_title, issue_description, payload.user_msg
)
verified_repos = await JiraFactory._verify_repos(
payload.issue_key, potential_repos, provider_handler
)
return JiraFactory._select_single_repo(
payload.issue_key, potential_repos, verified_repos
)
@staticmethod
async def create_view(
payload: JiraWebhookPayload,
workspace: JiraWorkspace,
user: JiraUser,
user_auth: UserAuth,
decrypted_api_key: str,
) -> JiraViewInterface:
"""Create appropriate Jira view based on the message and user state"""
"""Create a Jira view with repository already selected.
if not jira_user or not saas_user_auth or not jira_workspace:
raise StartingConvoException('User not authenticated with Jira integration')
This factory method:
1. Creates the view with payload and auth context
2. Fetches issue details (needed for repo inference)
3. Infers and selects the repository
conversation = await integration_store.get_user_conversations_by_issue_id(
job_context.issue_id, jira_user.id
If any step fails, an appropriate exception is raised with
a user-friendly message.
Args:
payload: Parsed webhook payload
workspace: The Jira workspace
user: The Jira user
user_auth: OpenHands user authentication
decrypted_api_key: Decrypted service account API key
Returns:
A JiraViewInterface with selected_repo populated
Raises:
StartingConvoException: If view creation fails
RepositoryNotFoundError: If repository cannot be determined
"""
logger.info(
'[Jira] Creating view',
extra={
'issue_key': payload.issue_key,
'event_type': payload.event_type.value,
},
)
if conversation:
logger.info(
f'[Jira] Found existing conversation for issue {job_context.issue_id}'
)
return JiraExistingConversationView(
job_context=job_context,
saas_user_auth=saas_user_auth,
jira_user=jira_user,
jira_workspace=jira_workspace,
selected_repo=None,
conversation_id=conversation.conversation_id,
)
return JiraNewConversationView(
job_context=job_context,
saas_user_auth=saas_user_auth,
jira_user=jira_user,
jira_workspace=jira_workspace,
selected_repo=None, # Will be set later after repo inference
conversation_id='', # Will be set when conversation is created
# Create the view
view = JiraNewConversationView(
payload=payload,
saas_user_auth=user_auth,
jira_user=user,
jira_workspace=workspace,
_decrypted_api_key=decrypted_api_key,
)
# Fetch issue details (needed for repo inference)
try:
issue_title, issue_description = await view.get_issue_details()
except StartingConvoException:
raise # Re-raise with original message
except Exception as e:
raise StartingConvoException(f'Failed to fetch issue details: {str(e)}')
# Infer and select repository
selected_repo = await JiraFactory._infer_repository(
payload=payload,
user_auth=user_auth,
issue_title=issue_title,
issue_description=issue_description,
)
view.selected_repo = selected_repo
logger.info(
'[Jira] View created successfully',
extra={
'issue_key': payload.issue_key,
'selected_repo': selected_repo,
},
)
return view

View File

@@ -19,6 +19,7 @@ from integrations.utils import (
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
filter_potential_repos_by_user_msg,
get_session_expired_message,
)
from jinja2 import Environment, FileSystemLoader
from server.auth.saas_user_auth import get_user_auth_from_keycloak_id
@@ -32,12 +33,16 @@ from openhands.core.logger import openhands_logger as logger
from openhands.integrations.provider import ProviderHandler
from openhands.integrations.service_types import Repository
from openhands.server.shared import server_config
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.server.user_auth.user_auth import UserAuth
from openhands.utils.http_session import httpx_verify_option
class JiraDcManager(Manager):
class JiraDcManager(Manager[JiraDcViewInterface]):
def __init__(self, token_manager: TokenManager):
self.token_manager = token_manager
self.integration_store = JiraDcIntegrationStore.get_instance()
@@ -348,7 +353,7 @@ class JiraDcManager(Manager):
logger.error(f'[Jira DC] Error in is_job_requested: {str(e)}')
return False
async def start_job(self, jira_dc_view: JiraDcViewInterface):
async def start_job(self, jira_dc_view: JiraDcViewInterface) -> None:
"""Start a Jira DC job/conversation."""
# Import here to prevent circular import
from server.conversation_callback_processor.jira_dc_callback_processor import (
@@ -397,6 +402,10 @@ class JiraDcManager(Manager):
logger.warning(f'[Jira DC] LLM authentication error: {str(e)}')
msg_info = f'Please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
except SessionExpiredError as e:
logger.warning(f'[Jira DC] Session expired: {str(e)}')
msg_info = get_session_expired_message()
except Exception as e:
logger.error(
f'[Jira DC] Unexpected error starting job: {str(e)}', exc_info=True
@@ -409,7 +418,7 @@ class JiraDcManager(Manager):
jira_dc_view.jira_dc_workspace.svc_acc_api_key
)
await self.send_message(
self.create_outgoing_message(msg=msg_info),
msg_info,
issue_key=jira_dc_view.job_context.issue_key,
base_api_url=jira_dc_view.job_context.base_api_url,
svc_acc_api_key=api_key,
@@ -447,12 +456,19 @@ class JiraDcManager(Manager):
return title, description
async def send_message(
self, message: Message, issue_key: str, base_api_url: str, svc_acc_api_key: str
self, message: str, issue_key: str, base_api_url: str, svc_acc_api_key: str
):
"""Send message/comment to Jira DC issue."""
"""Send message/comment to Jira DC issue.
Args:
message: The message content to send (plain text string)
issue_key: The Jira issue key (e.g., 'PROJ-123')
base_api_url: The base API URL for the Jira DC instance
svc_acc_api_key: Service account API key for authentication
"""
url = f'{base_api_url}/rest/api/2/issue/{issue_key}/comment'
headers = {'Authorization': f'Bearer {svc_acc_api_key}'}
data = {'body': message.message}
data = {'body': message}
async with httpx.AsyncClient(verify=httpx_verify_option()) as client:
response = await client.post(url, headers=headers, json=data)
response.raise_for_status()
@@ -472,7 +488,7 @@ class JiraDcManager(Manager):
try:
api_key = self.token_manager.decrypt_text(workspace.svc_acc_api_key)
await self.send_message(
self.create_outgoing_message(msg=error_msg),
error_msg,
issue_key=job_context.issue_key,
base_api_url=job_context.base_api_url,
svc_acc_api_key=api_key,
@@ -493,7 +509,7 @@ class JiraDcManager(Manager):
)
await self.send_message(
self.create_outgoing_message(msg=comment_msg),
comment_msg,
issue_key=jira_dc_view.job_context.issue_key,
base_api_url=jira_dc_view.job_context.base_api_url,
svc_acc_api_key=api_key,

View File

@@ -19,7 +19,7 @@ class JiraDcViewInterface(ABC):
conversation_id: str
@abstractmethod
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Get initial instructions for the conversation."""
pass

View File

@@ -36,7 +36,7 @@ class JiraDcNewConversationView(JiraDcViewInterface):
selected_repo: str | None
conversation_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
instructions_template = jinja_env.get_template('jira_dc_instructions.j2')
@@ -61,7 +61,7 @@ class JiraDcNewConversationView(JiraDcViewInterface):
provider_tokens = await self.saas_user_auth.get_provider_tokens()
user_secrets = await self.saas_user_auth.get_secrets()
instructions, user_msg = self._get_instructions(jinja_env)
instructions, user_msg = await self._get_instructions(jinja_env)
try:
agent_loop_info = await create_new_conversation(
@@ -113,7 +113,7 @@ class JiraDcExistingConversationView(JiraDcViewInterface):
selected_repo: str | None
conversation_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
user_msg_template = jinja_env.get_template('jira_dc_existing_conversation.j2')
@@ -155,6 +155,9 @@ class JiraDcExistingConversationView(JiraDcViewInterface):
self.conversation_id, conversation_init_data, user_id
)
if agent_loop_info.event_store is None:
raise StartingConvoException('Event store not available')
final_agent_observation = get_final_agent_observation(
agent_loop_info.event_store
)
@@ -167,7 +170,7 @@ class JiraDcExistingConversationView(JiraDcViewInterface):
if not agent_state or agent_state == AgentState.LOADING:
raise StartingConvoException('Conversation is still starting')
_, user_msg = self._get_instructions(jinja_env)
_, user_msg = await self._get_instructions(jinja_env)
user_message_event = MessageAction(content=user_msg)
await conversation_manager.send_event_to_conversation(
self.conversation_id, event_to_dict(user_message_event)

View File

@@ -16,6 +16,7 @@ from integrations.utils import (
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
filter_potential_repos_by_user_msg,
get_session_expired_message,
)
from jinja2 import Environment, FileSystemLoader
from server.auth.saas_user_auth import get_user_auth_from_keycloak_id
@@ -29,12 +30,16 @@ from openhands.core.logger import openhands_logger as logger
from openhands.integrations.provider import ProviderHandler
from openhands.integrations.service_types import Repository
from openhands.server.shared import server_config
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.server.user_auth.user_auth import UserAuth
from openhands.utils.http_session import httpx_verify_option
class LinearManager(Manager):
class LinearManager(Manager[LinearViewInterface]):
def __init__(self, token_manager: TokenManager):
self.token_manager = token_manager
self.integration_store = LinearIntegrationStore.get_instance()
@@ -338,7 +343,7 @@ class LinearManager(Manager):
logger.error(f'[Linear] Error in is_job_requested: {str(e)}')
return False
async def start_job(self, linear_view: LinearViewInterface):
async def start_job(self, linear_view: LinearViewInterface) -> None:
"""Start a Linear job/conversation."""
# Import here to prevent circular import
from server.conversation_callback_processor.linear_callback_processor import (
@@ -387,6 +392,10 @@ class LinearManager(Manager):
logger.warning(f'[Linear] LLM authentication error: {str(e)}')
msg_info = f'Please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
except SessionExpiredError as e:
logger.warning(f'[Linear] Session expired: {str(e)}')
msg_info = get_session_expired_message()
except Exception as e:
logger.error(
f'[Linear] Unexpected error starting job: {str(e)}', exc_info=True
@@ -399,7 +408,7 @@ class LinearManager(Manager):
linear_view.linear_workspace.svc_acc_api_key
)
await self.send_message(
self.create_outgoing_message(msg=msg_info),
msg_info,
linear_view.job_context.issue_id,
api_key,
)
@@ -464,8 +473,14 @@ class LinearManager(Manager):
return title, description
async def send_message(self, message: Message, issue_id: str, api_key: str):
"""Send message/comment to Linear issue."""
async def send_message(self, message: str, issue_id: str, api_key: str):
"""Send message/comment to Linear issue.
Args:
message: The message content to send (plain text string)
issue_id: The Linear issue ID to comment on
api_key: The Linear API key for authentication
"""
query = """
mutation CommentCreate($input: CommentCreateInput!) {
commentCreate(input: $input) {
@@ -476,7 +491,7 @@ class LinearManager(Manager):
}
}
"""
variables = {'input': {'issueId': issue_id, 'body': message.message}}
variables = {'input': {'issueId': issue_id, 'body': message}}
return await self._query_api(query, variables, api_key)
async def _send_error_comment(
@@ -489,9 +504,7 @@ class LinearManager(Manager):
try:
api_key = self.token_manager.decrypt_text(workspace.svc_acc_api_key)
await self.send_message(
self.create_outgoing_message(msg=error_msg), issue_id, api_key
)
await self.send_message(error_msg, issue_id, api_key)
except Exception as e:
logger.error(f'[Linear] Failed to send error comment: {str(e)}')
@@ -508,7 +521,7 @@ class LinearManager(Manager):
)
await self.send_message(
self.create_outgoing_message(msg=comment_msg),
comment_msg,
linear_view.job_context.issue_id,
api_key,
)

View File

@@ -19,7 +19,7 @@ class LinearViewInterface(ABC):
conversation_id: str
@abstractmethod
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Get initial instructions for the conversation."""
pass

View File

@@ -33,7 +33,7 @@ class LinearNewConversationView(LinearViewInterface):
selected_repo: str | None
conversation_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
instructions_template = jinja_env.get_template('linear_instructions.j2')
@@ -58,7 +58,7 @@ class LinearNewConversationView(LinearViewInterface):
provider_tokens = await self.saas_user_auth.get_provider_tokens()
user_secrets = await self.saas_user_auth.get_secrets()
instructions, user_msg = self._get_instructions(jinja_env)
instructions, user_msg = await self._get_instructions(jinja_env)
try:
agent_loop_info = await create_new_conversation(
@@ -110,7 +110,7 @@ class LinearExistingConversationView(LinearViewInterface):
selected_repo: str | None
conversation_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
user_msg_template = jinja_env.get_template('linear_existing_conversation.j2')
@@ -152,6 +152,9 @@ class LinearExistingConversationView(LinearViewInterface):
self.conversation_id, conversation_init_data, user_id
)
if agent_loop_info.event_store is None:
raise StartingConvoException('Event store not available')
final_agent_observation = get_final_agent_observation(
agent_loop_info.event_store
)
@@ -164,7 +167,7 @@ class LinearExistingConversationView(LinearViewInterface):
if not agent_state or agent_state == AgentState.LOADING:
raise StartingConvoException('Conversation is still starting')
_, user_msg = self._get_instructions(jinja_env)
_, user_msg = await self._get_instructions(jinja_env)
user_message_event = MessageAction(content=user_msg)
await conversation_manager.send_event_to_conversation(
self.conversation_id, event_to_dict(user_message_event)

View File

@@ -1,9 +1,13 @@
from abc import ABC, abstractmethod
from typing import Any, Generic, TypeVar
from integrations.models import Message, SourceType
# TypeVar for view types - each manager subclass specifies its own view type
ViewT = TypeVar('ViewT')
class Manager(ABC):
class Manager(ABC, Generic[ViewT]):
manager_type: SourceType
@abstractmethod
@@ -12,19 +16,21 @@ class Manager(ABC):
raise NotImplementedError
@abstractmethod
def send_message(self, message: Message):
"Send message to integration from Openhands server"
def send_message(self, message: str, *args: Any, **kwargs: Any):
"""Send message to integration from OpenHands server.
Args:
message: The message content to send (plain text string).
"""
raise NotImplementedError
@abstractmethod
async def is_job_requested(self, message: Message) -> bool:
"Confirm that a job is being requested"
raise NotImplementedError
async def start_job(self, view: ViewT) -> None:
"""Kick off a job with openhands agent.
@abstractmethod
def start_job(self):
"Kick off a job with openhands agent"
Args:
view: Integration-specific view object containing job context.
Each manager subclass accepts its own view type
(e.g., SlackViewInterface, JiraViewInterface, etc.)
"""
raise NotImplementedError
def create_outgoing_message(self, msg: str | dict, ephemeral: bool = False):
return Message(source=SourceType.OPENHANDS, message=msg, ephemeral=ephemeral)

View File

@@ -1,4 +1,5 @@
from enum import Enum
from typing import Any
from pydantic import BaseModel
@@ -16,8 +17,16 @@ class SourceType(str, Enum):
class Message(BaseModel):
"""Message model for incoming webhook payloads from integrations.
Note: This model is intended for INCOMING messages only.
For outgoing messages (e.g., sending comments to GitHub/GitLab),
pass strings directly to the send_message methods instead of
wrapping them in a Message object.
"""
source: SourceType
message: str | dict
message: dict[str, Any]
ephemeral: bool = False

View File

@@ -1,6 +1,6 @@
from openhands.app_server.user.user_context import UserContext
from openhands.app_server.user.user_models import UserInfo
from openhands.integrations.provider import PROVIDER_TOKEN_TYPE
from openhands.integrations.provider import PROVIDER_TOKEN_TYPE, ProviderHandler
from openhands.integrations.service_types import ProviderType
from openhands.sdk.secret import SecretSource, StaticSecret
from openhands.server.user_auth.user_auth import UserAuth
@@ -14,6 +14,7 @@ class ResolverUserContext(UserContext):
saas_user_auth: UserAuth,
):
self.saas_user_auth = saas_user_auth
self._provider_handler: ProviderHandler | None = None
async def get_user_id(self) -> str | None:
return await self.saas_user_auth.get_user_id()
@@ -29,20 +30,39 @@ class ResolverUserContext(UserContext):
return UserInfo(id=user_id)
async def get_authenticated_git_url(self, repository: str) -> str:
# This would need to be implemented based on the git provider tokens
# For now, return a basic HTTPS URL
return f'https://github.com/{repository}.git'
async def _get_provider_handler(self) -> ProviderHandler:
"""Get or create a ProviderHandler for git operations."""
if self._provider_handler is None:
provider_tokens = await self.saas_user_auth.get_provider_tokens()
if provider_tokens is None:
raise ValueError('No provider tokens available')
user_id = await self.saas_user_auth.get_user_id()
self._provider_handler = ProviderHandler(
provider_tokens=provider_tokens, external_auth_id=user_id
)
return self._provider_handler
async def get_authenticated_git_url(
self, repository: str, is_optional: bool = False
) -> str:
provider_handler = await self._get_provider_handler()
url = await provider_handler.get_authenticated_git_url(
repository, is_optional=is_optional
)
return url
async def get_latest_token(self, provider_type: ProviderType) -> str | None:
# Return the appropriate token from git_provider_tokens
# Return the appropriate token string from git_provider_tokens
provider_tokens = await self.saas_user_auth.get_provider_tokens()
if provider_tokens:
return provider_tokens.get(provider_type)
provider_token = provider_tokens.get(provider_type)
if provider_token and provider_token.token:
return provider_token.token.get_secret_value()
return None
async def get_provider_tokens(self) -> PROVIDER_TOKEN_TYPE | None:
async def get_provider_tokens(
self, as_env_vars: bool = False
) -> PROVIDER_TOKEN_TYPE | dict[str, str] | None:
return await self.saas_user_auth.get_provider_tokens()
async def get_secrets(self) -> dict[str, SecretSource]:

View File

@@ -0,0 +1,128 @@
"""Centralized error handling for Slack integration.
This module provides:
- SlackErrorCode: Unique error codes for traceability
- SlackError: Exception class for user-facing errors
- get_user_message(): Function to get user-facing messages for error codes
"""
import logging
from enum import Enum
from typing import Any
from integrations.utils import HOST_URL
logger = logging.getLogger(__name__)
class SlackErrorCode(Enum):
"""Unique error codes for traceability in logs and user messages."""
SESSION_EXPIRED = 'SLACK_ERR_001'
REDIS_STORE_FAILED = 'SLACK_ERR_002'
REDIS_RETRIEVE_FAILED = 'SLACK_ERR_003'
USER_NOT_AUTHENTICATED = 'SLACK_ERR_004'
PROVIDER_TIMEOUT = 'SLACK_ERR_005'
PROVIDER_AUTH_FAILED = 'SLACK_ERR_006'
LLM_AUTH_FAILED = 'SLACK_ERR_007'
MISSING_SETTINGS = 'SLACK_ERR_008'
UNEXPECTED_ERROR = 'SLACK_ERR_999'
class SlackError(Exception):
"""Exception for errors that should be communicated to the Slack user.
This exception is caught by the centralized error handler in SlackManager,
which logs the error and sends an appropriate message to the user.
Usage:
raise SlackError(SlackErrorCode.USER_NOT_AUTHENTICATED,
message_kwargs={'login_link': link})
"""
def __init__(
self,
code: SlackErrorCode,
message_kwargs: dict[str, Any] | None = None,
log_context: dict[str, Any] | None = None,
):
"""Initialize a SlackError.
Args:
code: The error code identifying the type of error
message_kwargs: Kwargs for formatting the user message
(e.g., {'login_link': '...'})
log_context: Additional context for structured logging
"""
self.code = code
self.message_kwargs = message_kwargs or {}
self.log_context = log_context or {}
super().__init__(f'{code.value}: {code.name}')
def get_user_message(self) -> str:
"""Get the user-facing message for this error."""
return get_user_message(self.code, **self.message_kwargs)
# Centralized user-facing messages
_USER_MESSAGES: dict[SlackErrorCode, str] = {
SlackErrorCode.SESSION_EXPIRED: (
'⏰ Your session has expired. '
'Please mention me again with your request to start a new conversation.'
),
SlackErrorCode.REDIS_STORE_FAILED: (
'⚠️ Something went wrong on our end (ref: {code}). '
'Please try again in a few moments.'
),
SlackErrorCode.REDIS_RETRIEVE_FAILED: (
'⚠️ Something went wrong on our end (ref: {code}). '
'Please try again in a few moments.'
),
SlackErrorCode.USER_NOT_AUTHENTICATED: (
'🔐 Please link your Slack account to OpenHands: '
'[Click here to Login]({login_link})'
),
SlackErrorCode.PROVIDER_TIMEOUT: (
'⏱️ The request timed out while connecting to your git provider. '
'Please try again.'
),
SlackErrorCode.PROVIDER_AUTH_FAILED: (
'🔐 Authentication with your git provider failed. '
f'Please re-login at [OpenHands Cloud]({HOST_URL}) and try again.'
),
SlackErrorCode.LLM_AUTH_FAILED: (
'@{username} please set a valid LLM API key in '
f'[OpenHands Cloud]({HOST_URL}) before starting a job.'
),
SlackErrorCode.MISSING_SETTINGS: (
'{username} please re-login into '
f'[OpenHands Cloud]({HOST_URL}) before starting a job.'
),
SlackErrorCode.UNEXPECTED_ERROR: (
'Uh oh! There was an unexpected error (ref: {code}). Please try again later.'
),
}
def get_user_message(error_code: SlackErrorCode, **kwargs) -> str:
"""Get a user-facing message for a given error code.
Args:
error_code: The error code to get a message for
**kwargs: Additional formatting arguments (e.g., username, login_link)
Returns:
Formatted user-facing message string
"""
msg = _USER_MESSAGES.get(
error_code, _USER_MESSAGES[SlackErrorCode.UNEXPECTED_ERROR]
)
try:
return msg.format(code=error_code.value, **kwargs)
except KeyError as e:
logger.warning(
f'Missing format key {e} in error message',
extra={'error_code': error_code.value},
)
# Return a generic error message with the code for debugging
return f'An error occurred (ref: {error_code.value}). Please try again later.'

View File

@@ -1,35 +1,49 @@
import re
from typing import Any
import jwt
from integrations.manager import Manager
from integrations.models import Message, SourceType
from integrations.slack.slack_types import SlackViewInterface, StartingConvoException
from integrations.slack.slack_errors import SlackError, SlackErrorCode
from integrations.slack.slack_types import (
SlackMessageView,
SlackViewInterface,
StartingConvoException,
)
from integrations.slack.slack_view import (
SlackFactory,
SlackNewConversationFromRepoFormView,
SlackNewConversationView,
SlackUnkownUserView,
SlackUpdateExistingConversationView,
)
from integrations.utils import (
HOST_URL,
OPENHANDS_RESOLVER_TEMPLATES_DIR,
get_session_expired_message,
infer_repo_from_message,
)
from integrations.v1_utils import get_saas_user_auth
from jinja2 import Environment, FileSystemLoader
from pydantic import SecretStr
from server.auth.saas_user_auth import SaasUserAuth
from server.constants import SLACK_CLIENT_ID
from server.utils.conversation_callback_utils import register_callback_processor
from slack_sdk.oauth import AuthorizeUrlGenerator
from slack_sdk.web.async_client import AsyncWebClient
from storage.database import session_maker
from sqlalchemy import select
from storage.database import a_session_maker
from storage.slack_user import SlackUser
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.provider import ProviderHandler
from openhands.integrations.service_types import Repository
from openhands.server.shared import config, server_config
from openhands.server.types import LLMAuthenticationError, MissingSettingsError
from openhands.integrations.service_types import (
AuthenticationError,
ProviderTimeoutError,
Repository,
)
from openhands.server.shared import config, server_config, sio
from openhands.server.types import (
LLMAuthenticationError,
MissingSettingsError,
SessionExpiredError,
)
from openhands.server.user_auth.user_auth import UserAuth
authorize_url_generator = AuthorizeUrlGenerator(
@@ -38,8 +52,14 @@ authorize_url_generator = AuthorizeUrlGenerator(
user_scopes=['search:read'],
)
# Key prefix for storing user messages in Redis during repo selection flow
SLACK_USER_MSG_KEY_PREFIX = 'slack_user_msg'
# Expiration time for stored user messages (5 minutes)
# Arbitrary timeout based on typical user attention span; may be tuned based on feedback
SLACK_USER_MSG_EXPIRATION = 300
class SlackManager(Manager):
class SlackManager(Manager[SlackViewInterface]):
def __init__(self, token_manager):
self.token_manager = token_manager
self.login_link = (
@@ -54,50 +74,148 @@ class SlackManager(Manager):
if message.source != SourceType.SLACK:
raise ValueError(f'Unexpected message source {message.source}')
async def _get_user_auth(self, keycloak_user_id: str) -> UserAuth:
offline_token = await self.token_manager.load_offline_token(keycloak_user_id)
if offline_token is None:
logger.info('no_offline_token_found')
user_auth = SaasUserAuth(
user_id=keycloak_user_id,
refresh_token=SecretStr(offline_token),
)
return user_auth
async def authenticate_user(
self, slack_user_id: str
) -> tuple[SlackUser | None, UserAuth | None]:
# We get the user and correlate them back to a user in OpenHands - if we can
slack_user = None
with session_maker() as session:
slack_user = (
session.query(SlackUser)
.filter(SlackUser.slack_user_id == slack_user_id)
.first()
async with a_session_maker() as session:
result = await session.execute(
select(SlackUser).where(SlackUser.slack_user_id == slack_user_id)
)
slack_user = result.scalar_one_or_none()
# slack_view.slack_to_openhands_user = slack_user # attach user auth info to view
saas_user_auth = None
if slack_user:
saas_user_auth = await self._get_user_auth(slack_user.keycloak_user_id)
saas_user_auth = await get_saas_user_auth(
slack_user.keycloak_user_id, self.token_manager
)
# slack_view.saas_user_auth = await self._get_user_auth(slack_view.slack_to_openhands_user.keycloak_user_id)
return slack_user, saas_user_auth
def _infer_repo_from_message(self, user_msg: str) -> str | None:
# Regular expression to match patterns like "OpenHands/OpenHands" or "deploy repo"
pattern = r'([a-zA-Z0-9_-]+/[a-zA-Z0-9_-]+)|([a-zA-Z0-9_-]+)(?=\s+repo)'
match = re.search(pattern, user_msg)
async def _store_user_msg_for_form(
self, message_ts: str, thread_ts: str | None, user_msg: str
) -> None:
"""Store user message in Redis for later retrieval when form is submitted.
if match:
repo = match.group(1) if match.group(1) else match.group(2)
return repo
This is needed because when a user selects a repo from the external_select
dropdown, Slack sends a separate interaction payload that doesn't include
the original user message.
return None
Args:
message_ts: The message timestamp (unique identifier)
thread_ts: The thread timestamp (if in a thread)
user_msg: The original user message to store
async def _get_repositories(self, user_auth: UserAuth) -> list[Repository]:
Raises:
SlackError: If storage fails (REDIS_STORE_FAILED)
"""
key = f'{SLACK_USER_MSG_KEY_PREFIX}:{message_ts}:{thread_ts}'
try:
redis = sio.manager.redis
await redis.set(key, user_msg, ex=SLACK_USER_MSG_EXPIRATION)
logger.info(
'slack_stored_user_msg',
extra={
'message_ts': message_ts,
'thread_ts': thread_ts,
'key': key,
},
)
except Exception as e:
logger.error(
'slack_store_user_msg_failed',
extra={
'message_ts': message_ts,
'thread_ts': thread_ts,
'key': key,
'error': str(e),
},
)
raise SlackError(
SlackErrorCode.REDIS_STORE_FAILED,
log_context={'message_ts': message_ts, 'thread_ts': thread_ts},
)
async def _retrieve_user_msg_for_form(
self, message_ts: str, thread_ts: str | None
) -> str:
"""Retrieve stored user message from Redis.
Args:
message_ts: The message timestamp
thread_ts: The thread timestamp (if in a thread)
Returns:
The stored user message
Raises:
SlackError: If retrieval fails (REDIS_RETRIEVE_FAILED) or message
not found (SESSION_EXPIRED)
"""
key = f'{SLACK_USER_MSG_KEY_PREFIX}:{message_ts}:{thread_ts}'
try:
redis = sio.manager.redis
user_msg = await redis.get(key)
if user_msg:
# Redis returns bytes, decode to string
if isinstance(user_msg, bytes):
user_msg = user_msg.decode('utf-8')
logger.info(
'slack_retrieved_user_msg',
extra={
'message_ts': message_ts,
'thread_ts': thread_ts,
'key': key,
},
)
return user_msg
else:
logger.warning(
'slack_user_msg_not_found',
extra={
'message_ts': message_ts,
'thread_ts': thread_ts,
'key': key,
},
)
raise SlackError(
SlackErrorCode.SESSION_EXPIRED,
log_context={'message_ts': message_ts, 'thread_ts': thread_ts},
)
except SlackError:
raise
except Exception as e:
logger.error(
'slack_retrieve_user_msg_failed',
extra={
'message_ts': message_ts,
'thread_ts': thread_ts,
'key': key,
'error': str(e),
},
)
raise SlackError(
SlackErrorCode.REDIS_RETRIEVE_FAILED,
log_context={'message_ts': message_ts, 'thread_ts': thread_ts},
)
async def _search_repositories(
self, user_auth: UserAuth, query: str = '', per_page: int = 100
) -> list[Repository]:
"""Search repositories for a user with optional query filtering.
Args:
user_auth: The user's authentication context
query: Search query to filter repositories (empty string returns all)
per_page: Maximum number of results to return
Returns:
List of matching Repository objects
"""
provider_tokens = await user_auth.get_provider_tokens()
if provider_tokens is None:
return []
@@ -108,31 +226,33 @@ class SlackManager(Manager):
external_auth_token=access_token,
external_auth_id=user_id,
)
repos: list[Repository] = await client.get_repositories(
'pushed', server_config.app_mode, None, None, None, None
repos: list[Repository] = await client.search_repositories(
selected_provider=None,
query=query,
per_page=per_page,
sort='pushed',
order='desc',
app_mode=server_config.app_mode,
)
return repos
def _generate_repo_selection_form(
self, repo_list: list[Repository], message_ts: str, thread_ts: str | None
):
options = [
{
'text': {'type': 'plain_text', 'text': 'No Repository'},
'value': '-',
}
]
options.extend(
{
'text': {
'type': 'plain_text',
'text': repo.full_name,
},
'value': repo.full_name,
}
for repo in repo_list
)
self, message_ts: str, thread_ts: str | None
) -> list[dict[str, Any]]:
"""Generate a repo selection form using external_select for dynamic loading.
This uses Slack's external_select element which allows:
- Type-ahead search for repositories
- Dynamic loading of options from an external endpoint
- Support for users with many repositories (no 100 option limit)
Args:
message_ts: The message timestamp for tracking
thread_ts: The thread timestamp if in a thread
Returns:
List of Slack Block Kit blocks for the selection form
"""
return [
{
'type': 'header',
@@ -142,160 +262,395 @@ class SlackManager(Manager):
'emoji': True,
},
},
{
'type': 'section',
'text': {
'type': 'mrkdwn',
'text': 'Type to search your repositories:',
},
},
{
'type': 'actions',
'elements': [
{
'type': 'static_select',
'type': 'external_select',
'action_id': f'repository_select:{message_ts}:{thread_ts}',
'options': options,
'placeholder': {
'type': 'plain_text',
'text': 'Search repositories...',
},
'min_query_length': 0, # Load initial options immediately
}
],
},
]
def filter_potential_repos_by_user_msg(
self, user_msg: str, user_repos: list[Repository]
) -> tuple[bool, list[Repository]]:
inferred_repo = self._infer_repo_from_message(user_msg)
if not inferred_repo:
return False, user_repos[0:99]
def _build_repo_options(self, repos: list[Repository]) -> list[dict[str, Any]]:
"""Build Slack options list from repositories.
final_repos = []
for repo in user_repos:
if inferred_repo.lower() in repo.full_name.lower():
final_repos.append(repo)
Always includes a "No Repository" option at the top, followed by up to 99
repositories (Slack has a 100 option limit for external_select).
# no repos matched, return original list
if len(final_repos) == 0:
return False, user_repos[0:99]
Args:
repos: List of Repository objects
# Found exact match
elif len(final_repos) == 1:
return True, final_repos
Returns:
List of Slack option objects
"""
options: list[dict[str, Any]] = [
{
'text': {'type': 'plain_text', 'text': 'No Repository'},
'value': '-',
}
]
options.extend(
{
'text': {
'type': 'plain_text',
'text': repo.full_name[:75], # Slack has 75 char limit for text
},
'value': repo.full_name,
}
for repo in repos[:99] # Leave room for "No Repository" option
)
return options
# Found partial matches
return False, final_repos[0:99]
async def search_repos_for_slack(
self, user_auth: UserAuth, query: str, per_page: int = 20
) -> list[dict[str, Any]]:
"""Public API for repository search with formatted Slack options.
This method searches for repositories and formats the results as Slack
external_select options.
Args:
user_auth: The user's authentication context
query: Search query to filter repositories (empty string returns all)
per_page: Maximum number of results to return (default: 20)
Returns:
List of Slack option objects ready for external_select response
"""
repos = await self._search_repositories(
user_auth, query=query, per_page=per_page
)
return self._build_repo_options(repos)
async def receive_message(self, message: Message):
"""Process an incoming Slack message.
This is the single entry point for all Slack message processing.
All SlackErrors raised during processing are caught and handled here,
sending appropriate error messages to the user.
"""
self._confirm_incoming_source_type(message)
try:
slack_view = await self._process_message(message)
if slack_view and await self.is_job_requested(message, slack_view):
await self.start_job(slack_view)
except SlackError as e:
await self.handle_slack_error(message.message, e)
except Exception as e:
logger.exception(
'slack_unexpected_error',
extra={'error': str(e), **message.message},
)
await self.handle_slack_error(
message.message,
SlackError(SlackErrorCode.UNEXPECTED_ERROR),
)
async def receive_form_interaction(self, slack_payload: dict):
"""Process a Slack form interaction (repository selection).
This handles the block_actions payload when a user selects a repository
from the dropdown form. It retrieves the original user message from Redis
and delegates to receive_message for processing.
Args:
slack_payload: The raw Slack interaction payload
"""
# Extract fields from the Slack interaction payload
selected_repository = slack_payload['actions'][0]['selected_option']['value']
if selected_repository == '-':
selected_repository = None
slack_user_id = slack_payload['user']['id']
channel_id = slack_payload['container']['channel_id']
team_id = slack_payload['team']['id']
# Get original message_ts and thread_ts from action_id
attribs = slack_payload['actions'][0]['action_id'].split('repository_select:')[
-1
]
message_ts, thread_ts = attribs.split(':')
thread_ts = None if thread_ts == 'None' else thread_ts
# Build partial payload for error handling during Redis retrieval
payload = {
'team_id': team_id,
'channel_id': channel_id,
'slack_user_id': slack_user_id,
'message_ts': message_ts,
'thread_ts': thread_ts,
}
# Retrieve the original user message from Redis
try:
user_msg = await self._retrieve_user_msg_for_form(message_ts, thread_ts)
except SlackError as e:
await self.handle_slack_error(payload, e)
return
except Exception as e:
logger.exception(
'slack_unexpected_error',
extra={'error': str(e), **payload},
)
await self.handle_slack_error(
payload, SlackError(SlackErrorCode.UNEXPECTED_ERROR)
)
return
# Complete the payload and delegate to receive_message
payload['selected_repo'] = selected_repository
payload['user_msg'] = user_msg
message = Message(source=SourceType.SLACK, message=payload)
await self.receive_message(message)
async def _process_message(self, message: Message) -> SlackViewInterface | None:
"""Process message and return view if authenticated, or raise SlackError.
Returns:
SlackViewInterface if user is authenticated and ready to proceed,
None if processing should stop (but no error).
Raises:
SlackError: If user is not authenticated or other recoverable error.
"""
slack_user, saas_user_auth = await self.authenticate_user(
slack_user_id=message.message['slack_user_id']
)
try:
slack_view = SlackFactory.create_slack_view_from_payload(
message, slack_user, saas_user_auth
slack_view = await SlackFactory.create_slack_view_from_payload(
message, slack_user, saas_user_auth
)
# Check if this is an unauthenticated user (SlackMessageView but not SlackViewInterface)
if not isinstance(slack_view, SlackViewInterface):
login_link = self._generate_login_link_with_state(message)
raise SlackError(
SlackErrorCode.USER_NOT_AUTHENTICATED,
message_kwargs={'login_link': login_link},
log_context=slack_view.to_log_context(),
)
except Exception as e:
return slack_view
def _generate_login_link_with_state(self, message: Message) -> str:
"""Generate OAuth login link with message state encoded."""
jwt_secret = config.jwt_secret
if not jwt_secret:
raise ValueError('Must configure jwt_secret')
state = jwt.encode(
message.message, jwt_secret.get_secret_value(), algorithm='HS256'
)
return authorize_url_generator.generate(state)
async def handle_slack_error(self, payload: dict, error: SlackError) -> None:
"""Handle a SlackError by logging and sending user message.
This is the centralized error handler for all SlackErrors, used by both
the manager and routes.
Args:
payload: The Slack payload dict containing channel/user info
error: The SlackError to handle
"""
# Create a minimal view for sending the error message
view = await SlackMessageView.from_payload(
payload, self._get_slack_team_store()
)
if not view:
logger.error(
f'[Slack]: Failed to create slack view: {e}',
exc_info=True,
stack_info=True,
'slack_error_no_view',
extra={
'error_code': error.code.value,
**error.log_context,
},
)
return
if isinstance(slack_view, SlackUnkownUserView):
jwt_secret = config.jwt_secret
if not jwt_secret:
raise ValueError('Must configure jwt_secret')
state = jwt.encode(
message.message, jwt_secret.get_secret_value(), algorithm='HS256'
)
link = authorize_url_generator.generate(state)
msg = self.login_link.format(link)
# Log the error
log_level = (
'exception' if error.code == SlackErrorCode.UNEXPECTED_ERROR else 'warning'
)
log_data = {
'error_code': error.code.value,
**view.to_log_context(),
**error.log_context,
}
getattr(logger, log_level)(
f'slack_error_{error.code.name.lower()}', extra=log_data
)
logger.info('slack_not_yet_authenticated')
await self.send_message(
self.create_outgoing_message(msg, ephemeral=True), slack_view
)
return
# Send user-facing message
await self.send_message(error.get_user_message(), view, ephemeral=True)
if not await self.is_job_requested(message, slack_view):
return
def _get_slack_team_store(self):
"""Get the SlackTeamStore instance (lazy import to avoid circular deps)."""
from storage.slack_team_store import SlackTeamStore
await self.start_job(slack_view)
return SlackTeamStore.get_instance()
async def send_message(self, message: Message, slack_view: SlackViewInterface):
async def send_message(
self,
message: str | dict[str, Any],
slack_view: SlackMessageView,
ephemeral: bool = False,
):
"""Send a message to Slack.
Args:
message: The message content. Can be a string (for simple text) or
a dict with 'text' and 'blocks' keys (for structured messages).
slack_view: The Slack view object containing channel/thread info.
Can be either SlackMessageView (for unauthenticated users)
or SlackViewInterface (for authenticated users).
ephemeral: If True, send as an ephemeral message visible only to the user.
"""
client = AsyncWebClient(token=slack_view.bot_access_token)
if message.ephemeral and isinstance(message.message, str):
if ephemeral and isinstance(message, str):
await client.chat_postEphemeral(
channel=slack_view.channel_id,
markdown_text=message.message,
markdown_text=message,
user=slack_view.slack_user_id,
thread_ts=slack_view.thread_ts,
)
elif message.ephemeral and isinstance(message.message, dict):
elif ephemeral and isinstance(message, dict):
await client.chat_postEphemeral(
channel=slack_view.channel_id,
user=slack_view.slack_user_id,
thread_ts=slack_view.thread_ts,
text=message.message['text'],
blocks=message.message['blocks'],
text=message['text'],
blocks=message['blocks'],
)
else:
await client.chat_postMessage(
channel=slack_view.channel_id,
markdown_text=message.message,
markdown_text=message,
thread_ts=slack_view.message_ts,
)
async def _try_verify_inferred_repo(
self, slack_view: SlackNewConversationView
) -> bool:
"""Try to infer and verify a repository from the user's message.
Returns:
True if a valid repo was found and verified, False otherwise
"""
user = slack_view.slack_to_openhands_user
inferred_repos = infer_repo_from_message(slack_view.user_msg)
if len(inferred_repos) != 1:
return False
inferred_repo = inferred_repos[0]
logger.info(
f'[Slack] Verifying inferred repo "{inferred_repo}" '
f'for user {user.slack_display_name} (id={slack_view.saas_user_auth.get_user_id()})'
)
try:
provider_tokens = await slack_view.saas_user_auth.get_provider_tokens()
if not provider_tokens:
return False
access_token = await slack_view.saas_user_auth.get_access_token()
user_id = await slack_view.saas_user_auth.get_user_id()
provider_handler = ProviderHandler(
provider_tokens=provider_tokens,
external_auth_token=access_token,
external_auth_id=user_id,
)
repo = await provider_handler.verify_repo_provider(inferred_repo)
slack_view.selected_repo = repo.full_name
return True
except (AuthenticationError, ProviderTimeoutError) as e:
logger.info(
f'[Slack] Could not verify repo "{inferred_repo}": {e}. '
f'Showing repository selector.'
)
return False
async def _show_repo_selection_form(
self, slack_view: SlackNewConversationView
) -> None:
"""Display the repository selection form to the user.
Raises:
SlackError: If storing the user message fails (REDIS_STORE_FAILED)
"""
user = slack_view.slack_to_openhands_user
logger.info(
'render_repository_selector',
extra={
'slack_user_id': user.slack_user_id,
'keycloak_user_id': user.keycloak_user_id,
'message_ts': slack_view.message_ts,
'thread_ts': slack_view.thread_ts,
},
)
# Store the user message for later retrieval - raises SlackError on failure
await self._store_user_msg_for_form(
slack_view.message_ts, slack_view.thread_ts, slack_view.user_msg
)
repo_selection_msg = {
'text': 'Choose a Repository:',
'blocks': self._generate_repo_selection_form(
slack_view.message_ts, slack_view.thread_ts
),
}
await self.send_message(repo_selection_msg, slack_view, ephemeral=True)
async def is_job_requested(
self, message: Message, slack_view: SlackViewInterface
) -> bool:
"""
A job is always request we only receive webhooks for events associated with the slack bot
This method really just checks
1. Is the user is authenticated
2. Do we have the necessary information to start a job (either by inferring the selected repo, otherwise asking the user)
"""
"""Determine if a job should be started based on the current context.
# Infer repo from user message is not needed; user selected repo from the form or is updating existing convo
This method checks:
1. If the view type allows immediate job start
2. If a repo can be inferred and verified from the message
3. Otherwise shows the repo selection form
Args:
slack_view: Must be a SlackViewType (authenticated view that can start jobs)
Returns:
True if job should start, False if waiting for user input
"""
# Check if view type allows immediate start
if isinstance(slack_view, SlackUpdateExistingConversationView):
return True
elif isinstance(slack_view, SlackNewConversationFromRepoFormView):
if isinstance(slack_view, SlackNewConversationFromRepoFormView):
return True
elif isinstance(slack_view, SlackNewConversationView):
user = slack_view.slack_to_openhands_user
user_repos: list[Repository] = await self._get_repositories(
slack_view.saas_user_auth
)
match, repos = self.filter_potential_repos_by_user_msg(
slack_view.user_msg, user_repos
)
# User mentioned a matching repo is their message, start job without repo selection form
if match:
slack_view.selected_repo = repos[0].full_name
# For new conversations, try to infer/verify repo or show selection form
if isinstance(slack_view, SlackNewConversationView):
if await self._try_verify_inferred_repo(slack_view):
return True
await self._show_repo_selection_form(slack_view)
logger.info(
'render_repository_selector',
extra={
'slack_user_id': user,
'keycloak_user_id': user.keycloak_user_id,
'message_ts': slack_view.message_ts,
'thread_ts': slack_view.thread_ts,
},
)
return False
repo_selection_msg = {
'text': 'Choose a Repository:',
'blocks': self._generate_repo_selection_form(
repos, slack_view.message_ts, slack_view.thread_ts
),
}
await self.send_message(
self.create_outgoing_message(repo_selection_msg, ephemeral=True),
slack_view,
)
return False
return True
async def start_job(self, slack_view: SlackViewInterface):
async def start_job(self, slack_view: SlackViewInterface) -> None:
# Importing here prevents circular import
from server.conversation_callback_processor.slack_callback_processor import (
SlackCallbackProcessor,
@@ -303,7 +658,7 @@ class SlackManager(Manager):
try:
msg_info = None
user_info: SlackUser = slack_view.slack_to_openhands_user
user_info = slack_view.slack_to_openhands_user
try:
logger.info(
f'[Slack] Starting job for user {user_info.slack_display_name} (id={user_info.slack_user_id})',
@@ -317,10 +672,15 @@ class SlackManager(Manager):
f'[Slack] Created conversation {conversation_id} for user {user_info.slack_display_name}'
)
if not isinstance(slack_view, SlackUpdateExistingConversationView):
# Only add SlackCallbackProcessor for new conversations (not updates) and non-v1 conversations
if (
not isinstance(slack_view, SlackUpdateExistingConversationView)
and not slack_view.v1_enabled
):
# We don't re-subscribe for follow up messages from slack.
# Summaries are generated for every messages anyways, we only need to do
# this subscription once for the event which kicked off the job.
processor = SlackCallbackProcessor(
slack_user_id=slack_view.slack_user_id,
channel_id=slack_view.channel_id,
@@ -335,6 +695,14 @@ class SlackManager(Manager):
logger.info(
f'[Slack] Created callback processor for conversation {conversation_id}'
)
elif isinstance(slack_view, SlackUpdateExistingConversationView):
logger.info(
f'[Slack] Skipping callback processor for existing conversation update {conversation_id}'
)
elif slack_view.v1_enabled:
logger.info(
f'[Slack] Skipping callback processor for v1 conversation {conversation_id}'
)
msg_info = slack_view.get_response_msg()
@@ -352,12 +720,20 @@ class SlackManager(Manager):
msg_info = f'@{user_info.slack_display_name} please set a valid LLM API key in [OpenHands Cloud]({HOST_URL}) before starting a job.'
except SessionExpiredError as e:
logger.warning(
f'[Slack] Session expired for user {user_info.slack_display_name}: {str(e)}'
)
msg_info = get_session_expired_message(user_info.slack_display_name)
except StartingConvoException as e:
msg_info = str(e)
await self.send_message(self.create_outgoing_message(msg_info), slack_view)
await self.send_message(msg_info, slack_view)
except Exception:
logger.exception('[Slack]: Error starting job')
msg = 'Uh oh! There was an unexpected error starting the job :('
await self.send_message(self.create_outgoing_message(msg), slack_view)
await self.send_message(
'Uh oh! There was an unexpected error starting the job :(', slack_view
)

View File

@@ -1,4 +1,5 @@
from abc import ABC, abstractmethod
from dataclasses import dataclass
from integrations.types import SummaryExtractionTracker
from jinja2 import Environment
@@ -7,34 +8,120 @@ from storage.slack_user import SlackUser
from openhands.server.user_auth.user_auth import UserAuth
class SlackViewInterface(SummaryExtractionTracker, ABC):
@dataclass
class SlackMessageView:
"""Minimal view for sending messages to Slack.
This class contains only the fields needed to send messages,
without requiring user authentication. Can be used directly for
simple message operations or as a base class for more complex views.
"""
bot_access_token: str
user_msg: str | None
slack_user_id: str
slack_to_openhands_user: SlackUser | None
saas_user_auth: UserAuth | None
channel_id: str
message_ts: str
thread_ts: str | None
team_id: str
def to_log_context(self) -> dict:
"""Return dict suitable for structured logging."""
return {
'slack_channel_id': self.channel_id,
'slack_user_id': self.slack_user_id,
'slack_team_id': self.team_id,
'slack_thread_ts': self.thread_ts,
'slack_message_ts': self.message_ts,
}
@classmethod
async def from_payload(
cls,
payload: dict,
slack_team_store,
) -> 'SlackMessageView | None':
"""Create a view from a raw Slack payload.
This factory method handles the various payload formats from different
Slack interactions (events, form submissions, block suggestions).
Args:
payload: Raw Slack payload dictionary
slack_team_store: Store for retrieving bot tokens
Returns:
SlackMessageView if all required fields are available,
None if required fields are missing or bot token unavailable.
"""
from openhands.core.logger import openhands_logger as logger
team_id = payload.get('team', {}).get('id') or payload.get('team_id')
channel_id = (
payload.get('container', {}).get('channel_id')
or payload.get('channel', {}).get('id')
or payload.get('channel_id')
)
user_id = payload.get('user', {}).get('id') or payload.get('slack_user_id')
message_ts = payload.get('message_ts', '')
thread_ts = payload.get('thread_ts')
if not team_id or not channel_id or not user_id:
logger.warning(
'slack_message_view_from_payload_missing_fields',
extra={
'has_team_id': bool(team_id),
'has_channel_id': bool(channel_id),
'has_user_id': bool(user_id),
'payload_keys': list(payload.keys()),
},
)
return None
bot_token = await slack_team_store.get_team_bot_token(team_id)
if not bot_token:
logger.warning(
'slack_message_view_from_payload_no_bot_token',
extra={'team_id': team_id},
)
return None
return cls(
bot_access_token=bot_token,
slack_user_id=user_id,
channel_id=channel_id,
message_ts=message_ts,
thread_ts=thread_ts,
team_id=team_id,
)
class SlackViewInterface(SlackMessageView, SummaryExtractionTracker, ABC):
"""Interface for authenticated Slack views that can create conversations.
All fields are required (non-None) because this interface is only used
for users who have linked their Slack account to OpenHands.
Inherits from SlackMessageView:
bot_access_token, slack_user_id, channel_id, message_ts, thread_ts, team_id
"""
user_msg: str
slack_to_openhands_user: SlackUser
saas_user_auth: UserAuth
selected_repo: str | None
should_extract: bool
send_summary_instruction: bool
conversation_id: str
team_id: str
v1_enabled: bool
@abstractmethod
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"Instructions passed when conversation is first initialized"
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
pass
@abstractmethod
async def create_or_update_conversation(self, jinja_env: Environment):
"Create a new conversation"
pass
@abstractmethod
def get_callback_id(self) -> str:
"Unique callback id for subscribription made to EventStream for fetching agent summary"
"""Create a new conversation"""
pass
@abstractmethod
@@ -43,6 +130,4 @@ class SlackViewInterface(SummaryExtractionTracker, ABC):
class StartingConvoException(Exception):
"""
Raised when trying to send message to a conversation that's is still starting up
"""
"""Raised when trying to send message to a conversation that is still starting up."""

View File

@@ -0,0 +1,268 @@
import logging
from uuid import UUID
import httpx
from integrations.utils import get_summary_instruction
from integrations.v1_utils import handle_callback_error
from pydantic import Field
from slack_sdk import WebClient
from storage.slack_team_store import SlackTeamStore
from openhands.agent_server.models import AskAgentRequest, AskAgentResponse
from openhands.app_server.event_callback.event_callback_models import (
EventCallback,
EventCallbackProcessor,
)
from openhands.app_server.event_callback.event_callback_result_models import (
EventCallbackResult,
EventCallbackResultStatus,
)
from openhands.app_server.event_callback.util import (
ensure_conversation_found,
ensure_running_sandbox,
get_agent_server_url_from_sandbox,
)
from openhands.sdk import Event
from openhands.sdk.event import ConversationStateUpdateEvent
_logger = logging.getLogger(__name__)
class SlackV1CallbackProcessor(EventCallbackProcessor):
"""Callback processor for Slack V1 integrations."""
slack_view_data: dict[str, str | None] = Field(default_factory=dict)
async def __call__(
self,
conversation_id: UUID,
callback: EventCallback,
event: Event,
) -> EventCallbackResult | None:
"""Process events for Slack V1 integration."""
# Only handle ConversationStateUpdateEvent
if not isinstance(event, ConversationStateUpdateEvent):
return None
# Only act when execution has finished
if not (event.key == 'execution_status' and event.value == 'finished'):
return None
_logger.info('[Slack V1] Callback agent state was %s', event)
try:
summary = await self._request_summary(conversation_id)
await self._post_summary_to_slack(summary)
return EventCallbackResult(
status=EventCallbackResultStatus.SUCCESS,
event_callback_id=callback.id,
event_id=event.id,
conversation_id=conversation_id,
detail=summary,
)
except Exception as e:
await handle_callback_error(
error=e,
conversation_id=conversation_id,
service_name='Slack',
service_logger=_logger,
can_post_error=True, # Slack always attempts to post errors
post_error_func=self._post_summary_to_slack,
)
return EventCallbackResult(
status=EventCallbackResultStatus.ERROR,
event_callback_id=callback.id,
event_id=event.id,
conversation_id=conversation_id,
detail=str(e),
)
# -------------------------------------------------------------------------
# Slack helpers
# -------------------------------------------------------------------------
async def _get_bot_access_token(self) -> str | None:
team_id = self.slack_view_data.get('team_id')
if team_id is None:
return None
slack_team_store = SlackTeamStore.get_instance()
bot_access_token = await slack_team_store.get_team_bot_token(team_id)
return bot_access_token
async def _post_summary_to_slack(self, summary: str) -> None:
"""Post a summary message to the configured Slack channel."""
bot_access_token = await self._get_bot_access_token()
if not bot_access_token:
raise RuntimeError('Missing Slack bot access token')
channel_id = self.slack_view_data['channel_id']
thread_ts = self.slack_view_data.get('thread_ts') or self.slack_view_data.get(
'message_ts'
)
client = WebClient(token=bot_access_token)
try:
# Post the summary as a threaded reply
response = client.chat_postMessage(
channel=channel_id,
text=summary,
thread_ts=thread_ts,
unfurl_links=False,
unfurl_media=False,
)
if not response['ok']:
raise RuntimeError(
f"Slack API error: {response.get('error', 'Unknown error')}"
)
_logger.info(
'[Slack V1] Successfully posted summary to channel %s', channel_id
)
except Exception as e:
_logger.error('[Slack V1] Failed to post message to Slack: %s', e)
raise
# -------------------------------------------------------------------------
# Agent / sandbox helpers
# -------------------------------------------------------------------------
async def _ask_question(
self,
httpx_client: httpx.AsyncClient,
agent_server_url: str,
conversation_id: UUID,
session_api_key: str,
message_content: str,
) -> str:
"""Send a message to the agent server via the V1 API and return response text."""
send_message_request = AskAgentRequest(question=message_content)
url = (
f"{agent_server_url.rstrip('/')}"
f"/api/conversations/{conversation_id}/ask_agent"
)
headers = {'X-Session-API-Key': session_api_key}
payload = send_message_request.model_dump()
try:
response = await httpx_client.post(
url,
json=payload,
headers=headers,
timeout=30.0,
)
response.raise_for_status()
agent_response = AskAgentResponse.model_validate(response.json())
return agent_response.response
except httpx.HTTPStatusError as e:
error_detail = f'HTTP {e.response.status_code} error'
try:
error_body = e.response.text
if error_body:
error_detail += f': {error_body}'
except Exception: # noqa: BLE001
pass
_logger.error(
'[Slack V1] HTTP error sending message to %s: %s. '
'Request payload: %s. Response headers: %s',
url,
error_detail,
payload,
dict(e.response.headers),
exc_info=True,
)
raise Exception(f'Failed to send message to agent server: {error_detail}')
except httpx.TimeoutException:
error_detail = f'Request timeout after 30 seconds to {url}'
_logger.error(
'[Slack V1] %s. Request payload: %s',
error_detail,
payload,
exc_info=True,
)
raise Exception(error_detail)
except httpx.RequestError as e:
error_detail = f'Request error to {url}: {str(e)}'
_logger.error(
'[Slack V1] %s. Request payload: %s',
error_detail,
payload,
exc_info=True,
)
raise Exception(error_detail)
# -------------------------------------------------------------------------
# Summary orchestration
# -------------------------------------------------------------------------
async def _request_summary(self, conversation_id: UUID) -> str:
"""Ask the agent to produce a summary of its work and return the agent response.
NOTE: This method now returns a string (the agent server's response text)
and raises exceptions on errors. The wrapping into EventCallbackResult
is handled by __call__.
"""
# Import services within the method to avoid circular imports
from openhands.app_server.config import (
get_app_conversation_info_service,
get_httpx_client,
get_sandbox_service,
)
from openhands.app_server.services.injector import InjectorState
from openhands.app_server.user.specifiy_user_context import (
ADMIN,
USER_CONTEXT_ATTR,
)
# Create injector state for dependency injection
state = InjectorState()
setattr(state, USER_CONTEXT_ATTR, ADMIN)
async with (
get_app_conversation_info_service(state) as app_conversation_info_service,
get_sandbox_service(state) as sandbox_service,
get_httpx_client(state) as httpx_client,
):
# 1. Conversation lookup
app_conversation_info = ensure_conversation_found(
await app_conversation_info_service.get_app_conversation_info(
conversation_id
),
conversation_id,
)
# 2. Sandbox lookup + validation
sandbox = ensure_running_sandbox(
await sandbox_service.get_sandbox(app_conversation_info.sandbox_id),
app_conversation_info.sandbox_id,
)
assert (
sandbox.session_api_key is not None
), f'No session API key for sandbox: {sandbox.id}'
# 3. URL + instruction
agent_server_url = get_agent_server_url_from_sandbox(sandbox)
# Prepare message based on agent state
message_content = get_summary_instruction()
# Ask the agent and return the response text
return await self._ask_question(
httpx_client=httpx_client,
agent_server_url=agent_server_url,
conversation_id=conversation_id,
session_api_key=sandbox.session_api_key,
message_content=message_content,
)

View File

@@ -1,8 +1,21 @@
import asyncio
from dataclasses import dataclass
from uuid import UUID, uuid4
from integrations.models import Message
from integrations.slack.slack_types import SlackViewInterface, StartingConvoException
from integrations.utils import CONVERSATION_URL, get_final_agent_observation
from integrations.resolver_context import ResolverUserContext
from integrations.slack.slack_types import (
SlackMessageView,
SlackViewInterface,
StartingConvoException,
)
from integrations.slack.slack_v1_callback_processor import SlackV1CallbackProcessor
from integrations.utils import (
CONVERSATION_URL,
ENABLE_V1_SLACK_RESOLVER,
get_final_agent_observation,
get_user_v1_enabled_setting,
)
from jinja2 import Environment
from slack_sdk import WebClient
from storage.slack_conversation import SlackConversation
@@ -10,22 +23,34 @@ from storage.slack_conversation_store import SlackConversationStore
from storage.slack_team_store import SlackTeamStore
from storage.slack_user import SlackUser
from openhands.app_server.app_conversation.app_conversation_models import (
AppConversationStartRequest,
AppConversationStartTaskStatus,
SendMessageRequest,
)
from openhands.app_server.config import get_app_conversation_service
from openhands.app_server.sandbox.sandbox_models import SandboxStatus
from openhands.app_server.services.injector import InjectorState
from openhands.app_server.user.specifiy_user_context import USER_CONTEXT_ATTR
from openhands.core.logger import openhands_logger as logger
from openhands.core.schema.agent import AgentState
from openhands.events.action import MessageAction
from openhands.events.serialization.event import event_to_dict
from openhands.integrations.provider import ProviderHandler
from openhands.integrations.provider import ProviderHandler, ProviderType
from openhands.sdk import TextContent
from openhands.server.services.conversation_service import (
create_new_conversation,
setup_init_conversation_settings,
)
from openhands.server.shared import ConversationStoreImpl, config, conversation_manager
from openhands.server.user_auth.user_auth import UserAuth
from openhands.storage.data_models.conversation_metadata import ConversationTrigger
from openhands.utils.async_utils import GENERAL_TIMEOUT, call_async_from_sync
from openhands.storage.data_models.conversation_metadata import (
ConversationTrigger,
)
from openhands.utils.async_utils import GENERAL_TIMEOUT
# =================================================
# SECTION: Github view types
# SECTION: Slack view types
# =================================================
@@ -34,39 +59,14 @@ slack_conversation_store = SlackConversationStore.get_instance()
slack_team_store = SlackTeamStore.get_instance()
@dataclass
class SlackUnkownUserView(SlackViewInterface):
bot_access_token: str
user_msg: str | None
slack_user_id: str
slack_to_openhands_user: SlackUser | None
saas_user_auth: UserAuth | None
channel_id: str
message_ts: str
thread_ts: str | None
selected_repo: str | None
should_extract: bool
send_summary_instruction: bool
conversation_id: str
team_id: str
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
raise NotImplementedError
async def create_or_update_conversation(self, jinja_env: Environment):
raise NotImplementedError
def get_callback_id(self) -> str:
raise NotImplementedError
def get_response_msg(self) -> str:
raise NotImplementedError
async def is_v1_enabled_for_slack_resolver(user_id: str) -> bool:
return await get_user_v1_enabled_setting(user_id) and ENABLE_V1_SLACK_RESOLVER
@dataclass
class SlackNewConversationView(SlackViewInterface):
bot_access_token: str
user_msg: str | None
user_msg: str
slack_user_id: str
slack_to_openhands_user: SlackUser
saas_user_auth: UserAuth
@@ -78,6 +78,7 @@ class SlackNewConversationView(SlackViewInterface):
send_summary_instruction: bool
conversation_id: str
team_id: str
v1_enabled: bool
def _get_initial_prompt(self, text: str, blocks: list[dict]):
bot_id = self._get_bot_id(blocks)
@@ -95,9 +96,8 @@ class SlackNewConversationView(SlackViewInterface):
return block['user_id']
return ''
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"Instructions passed when conversation is first initialized"
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized"""
user_info: SlackUser = self.slack_to_openhands_user
messages = []
@@ -157,7 +157,7 @@ class SlackNewConversationView(SlackViewInterface):
'Attempting to start conversation without confirming selected repo from user'
)
async def save_slack_convo(self):
async def save_slack_convo(self, v1_enabled: bool = False):
if self.slack_to_openhands_user:
user_info: SlackUser = self.slack_to_openhands_user
@@ -167,27 +167,62 @@ class SlackNewConversationView(SlackViewInterface):
'channel_id': self.channel_id,
'conversation_id': self.conversation_id,
'keycloak_user_id': user_info.keycloak_user_id,
'org_id': user_info.org_id,
'parent_id': self.thread_ts or self.message_ts,
'v1_enabled': v1_enabled,
},
)
slack_conversation = SlackConversation(
conversation_id=self.conversation_id,
channel_id=self.channel_id,
keycloak_user_id=user_info.keycloak_user_id,
org_id=user_info.org_id,
parent_id=self.thread_ts
or self.message_ts, # conversations can start in a thread reply as well; we should always references the parent's (root level msg's) message ID
v1_enabled=v1_enabled,
)
await slack_conversation_store.create_slack_conversation(slack_conversation)
def _create_slack_v1_callback_processor(self) -> SlackV1CallbackProcessor:
"""Create a SlackV1CallbackProcessor for V1 conversation handling."""
return SlackV1CallbackProcessor(
slack_view_data={
'channel_id': self.channel_id,
'message_ts': self.message_ts,
'thread_ts': self.thread_ts,
'team_id': self.team_id,
'slack_user_id': self.slack_user_id,
}
)
async def create_or_update_conversation(self, jinja: Environment) -> str:
"""
Only creates a new conversation
"""
"""Only creates a new conversation"""
self._verify_necessary_values_are_set()
provider_tokens = await self.saas_user_auth.get_provider_tokens()
user_secrets = await self.saas_user_auth.get_secrets()
user_instructions, conversation_instructions = self._get_instructions(jinja)
# Check if V1 conversations are enabled for this user
self.v1_enabled = await is_v1_enabled_for_slack_resolver(
self.slack_to_openhands_user.keycloak_user_id
)
if self.v1_enabled:
# Use V1 app conversation service
await self._create_v1_conversation(jinja)
return self.conversation_id
else:
# Use existing V0 conversation service
await self._create_v0_conversation(jinja, provider_tokens, user_secrets)
return self.conversation_id
async def _create_v0_conversation(
self, jinja: Environment, provider_tokens, user_secrets
) -> None:
"""Create conversation using the legacy V0 system."""
user_instructions, conversation_instructions = await self._get_instructions(
jinja
)
# Determine git provider from repository
git_provider = None
@@ -213,11 +248,67 @@ class SlackNewConversationView(SlackViewInterface):
)
self.conversation_id = agent_loop_info.conversation_id
await self.save_slack_convo()
return self.conversation_id
logger.info(f'[Slack]: Created V0 conversation: {self.conversation_id}')
await self.save_slack_convo(v1_enabled=False)
def get_callback_id(self) -> str:
return f'slack_{self.channel_id}_{self.message_ts}'
async def _create_v1_conversation(self, jinja: Environment) -> None:
"""Create conversation using the new V1 app conversation system."""
user_instructions, conversation_instructions = await self._get_instructions(
jinja
)
# Create the initial message request
initial_message = SendMessageRequest(
role='user', content=[TextContent(text=user_instructions)]
)
# Create the Slack V1 callback processor
slack_callback_processor = self._create_slack_v1_callback_processor()
# Determine git provider from repository
git_provider = None
provider_tokens = await self.saas_user_auth.get_provider_tokens()
if self.selected_repo and provider_tokens:
provider_handler = ProviderHandler(provider_tokens)
repository = await provider_handler.verify_repo_provider(self.selected_repo)
git_provider = ProviderType(repository.git_provider.value)
# Get the app conversation service and start the conversation
injector_state = InjectorState()
# Create the V1 conversation start request with the callback processor
self.conversation_id = uuid4().hex
start_request = AppConversationStartRequest(
conversation_id=UUID(self.conversation_id),
system_message_suffix=conversation_instructions,
initial_message=initial_message,
selected_repository=self.selected_repo,
git_provider=git_provider,
title=f'Slack conversation from {self.slack_to_openhands_user.slack_display_name}',
trigger=ConversationTrigger.SLACK,
processors=[
slack_callback_processor
], # Pass the callback processor directly
)
# Set up the Slack user context for the V1 system
slack_user_context = ResolverUserContext(saas_user_auth=self.saas_user_auth)
setattr(injector_state, USER_CONTEXT_ATTR, slack_user_context)
async with get_app_conversation_service(
injector_state
) as app_conversation_service:
async for task in app_conversation_service.start_app_conversation(
start_request
):
if task.status == AppConversationStartTaskStatus.ERROR:
logger.error(f'Failed to start V1 conversation: {task.detail}')
raise RuntimeError(
f'Failed to start V1 conversation: {task.detail}'
)
logger.info(f'[Slack V1]: Created new conversation: {self.conversation_id}')
await self.save_slack_convo(v1_enabled=True)
def get_response_msg(self) -> str:
user_info: SlackUser = self.slack_to_openhands_user
@@ -237,7 +328,7 @@ class SlackNewConversationFromRepoFormView(SlackNewConversationView):
class SlackUpdateExistingConversationView(SlackNewConversationView):
slack_conversation: SlackConversation
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
client = WebClient(token=self.bot_access_token)
result = client.conversations_replies(
channel=self.channel_id,
@@ -254,32 +345,20 @@ class SlackUpdateExistingConversationView(SlackNewConversationView):
return user_message, ''
async def create_or_update_conversation(self, jinja: Environment) -> str:
"""
Send new user message to converation
"""
async def send_message_to_v0_conversation(self, jinja: Environment):
user_info: SlackUser = self.slack_to_openhands_user
saas_user_auth: UserAuth = self.saas_user_auth
user_id = user_info.keycloak_user_id
# Org management in the future will get rid of this
# For now, only user that created the conversation can send follow up messages to it
if user_id != self.slack_conversation.keycloak_user_id:
raise StartingConvoException(
f'{user_info.slack_display_name} is not authorized to send messages to this conversation.'
)
# Check if conversation has been deleted
# Update logic when soft delete is implemented
conversation_store = await ConversationStoreImpl.get_instance(config, user_id)
saas_user_auth: UserAuth = self.saas_user_auth
provider_tokens = await saas_user_auth.get_provider_tokens()
try:
conversation_store = await ConversationStoreImpl.get_instance(
config, user_id
)
await conversation_store.get_metadata(self.conversation_id)
except FileNotFoundError:
raise StartingConvoException('Conversation no longer exists.')
provider_tokens = await saas_user_auth.get_provider_tokens()
# Should we raise here if there are no provider tokens?
providers_set = list(provider_tokens.keys()) if provider_tokens else []
@@ -292,6 +371,9 @@ class SlackUpdateExistingConversationView(SlackNewConversationView):
self.conversation_id, conversation_init_data, user_id
)
if agent_loop_info.event_store is None:
raise StartingConvoException('Event store not available')
final_agent_observation = get_final_agent_observation(
agent_loop_info.event_store
)
@@ -304,12 +386,123 @@ class SlackUpdateExistingConversationView(SlackNewConversationView):
if not agent_state or agent_state == AgentState.LOADING:
raise StartingConvoException('Conversation is still starting')
user_msg, _ = self._get_instructions(jinja)
user_msg_action = MessageAction(content=user_msg)
instructions, _ = await self._get_instructions(jinja)
user_msg = MessageAction(content=instructions)
await conversation_manager.send_event_to_conversation(
self.conversation_id, event_to_dict(user_msg_action)
self.conversation_id, event_to_dict(user_msg)
)
async def send_message_to_v1_conversation(self, jinja: Environment):
"""Send a message to a v1 conversation using the agent server API."""
# Import services within the method to avoid circular imports
from openhands.agent_server.models import SendMessageRequest
from openhands.app_server.config import (
get_app_conversation_info_service,
get_httpx_client,
get_sandbox_service,
)
from openhands.app_server.event_callback.util import (
ensure_conversation_found,
get_agent_server_url_from_sandbox,
)
from openhands.app_server.services.injector import InjectorState
from openhands.app_server.user.specifiy_user_context import (
ADMIN,
USER_CONTEXT_ATTR,
)
# Create injector state for dependency injection
state = InjectorState()
setattr(state, USER_CONTEXT_ATTR, ADMIN)
async with (
get_app_conversation_info_service(state) as app_conversation_info_service,
get_sandbox_service(state) as sandbox_service,
get_httpx_client(state) as httpx_client,
):
# 1. Conversation lookup
app_conversation_info = ensure_conversation_found(
await app_conversation_info_service.get_app_conversation_info(
UUID(self.conversation_id)
),
UUID(self.conversation_id),
)
# 2. Sandbox lookup + validation
sandbox = await sandbox_service.get_sandbox(
app_conversation_info.sandbox_id
)
if sandbox and sandbox.status == SandboxStatus.PAUSED:
# Resume paused sandbox and wait for it to be running
logger.info('[Slack V1]: Attempting to resume paused sandbox')
await sandbox_service.resume_sandbox(app_conversation_info.sandbox_id)
# Wait for sandbox to be running (handles both fresh start and resume)
running_sandbox = await sandbox_service.wait_for_sandbox_running(
app_conversation_info.sandbox_id,
timeout=120,
poll_interval=2,
httpx_client=httpx_client,
)
assert (
running_sandbox.session_api_key is not None
), f'No session API key for sandbox: {running_sandbox.id}'
# 3. Get the agent server URL
agent_server_url = get_agent_server_url_from_sandbox(running_sandbox)
# 4. Prepare the message content
user_msg, _ = await self._get_instructions(jinja)
# 5. Create the message request
send_message_request = SendMessageRequest(
role='user', content=[TextContent(text=user_msg)], run=True
)
# 6. Send the message to the agent server
url = f"{agent_server_url.rstrip('/')}/api/conversations/{UUID(self.conversation_id)}/events"
headers = {'X-Session-API-Key': running_sandbox.session_api_key}
payload = send_message_request.model_dump()
try:
response = await httpx_client.post(
url,
json=payload,
headers=headers,
timeout=30.0,
)
response.raise_for_status()
except Exception as e:
logger.error(
'[Slack V1] Failed to send message to conversation %s: %s',
self.conversation_id,
str(e),
exc_info=True,
)
raise Exception(f'Failed to send message to v1 conversation: {str(e)}')
async def create_or_update_conversation(self, jinja: Environment) -> str:
"""Send new user message to converation"""
user_info: SlackUser = self.slack_to_openhands_user
user_id = user_info.keycloak_user_id
# Org management in the future will get rid of this
# For now, only user that created the conversation can send follow up messages to it
if user_id != self.slack_conversation.keycloak_user_id:
raise StartingConvoException(
f'{user_info.slack_display_name} is not authorized to send messages to this conversation.'
)
if self.slack_conversation.v1_enabled:
await self.send_message_to_v1_conversation(jinja)
else:
await self.send_message_to_v0_conversation(jinja)
return self.conversation_id
def get_response_msg(self):
@@ -337,11 +530,14 @@ class SlackFactory:
return None
# thread_ts in slack payloads in the parent's (root level msg's) message ID
if channel_id is None:
return None
return await slack_conversation_store.get_slack_conversation(
channel_id, thread_ts
)
def create_slack_view_from_payload(
@staticmethod
async def create_slack_view_from_payload(
message: Message, slack_user: SlackUser | None, saas_user_auth: UserAuth | None
):
payload = message.message
@@ -352,7 +548,7 @@ class SlackFactory:
team_id = payload['team_id']
user_msg = payload.get('user_msg')
bot_access_token = slack_team_store.get_team_bot_token(team_id)
bot_access_token = await slack_team_store.get_team_bot_token(team_id)
if not bot_access_token:
logger.error(
'Did not find slack team',
@@ -361,30 +557,30 @@ class SlackFactory:
'channel_id': channel_id,
},
)
raise Exception('Did not slack team')
raise Exception('Did not find slack team')
# Determine if this is a known slack user by openhands
if not slack_user or not saas_user_auth or not channel_id:
return SlackUnkownUserView(
# Return SlackMessageView (not SlackViewInterface) for unauthenticated users
if not slack_user or not saas_user_auth or not channel_id or not message_ts:
return SlackMessageView(
bot_access_token=bot_access_token,
user_msg=user_msg,
slack_user_id=slack_user_id,
slack_to_openhands_user=slack_user,
saas_user_auth=saas_user_auth,
channel_id=channel_id,
message_ts=message_ts,
channel_id=channel_id or '',
message_ts=message_ts or '',
thread_ts=thread_ts,
selected_repo=None,
should_extract=False,
send_summary_instruction=False,
conversation_id='',
team_id=team_id,
)
conversation: SlackConversation | None = call_async_from_sync(
SlackFactory.determine_if_updating_existing_conversation,
GENERAL_TIMEOUT,
message,
# At this point, we've verified slack_user, saas_user_auth, channel_id, and message_ts are set
# user_msg should always be present in Slack payloads
if not user_msg:
raise ValueError('user_msg is required but was not provided in payload')
assert channel_id is not None
assert message_ts is not None
conversation = await asyncio.wait_for(
SlackFactory.determine_if_updating_existing_conversation(message),
timeout=GENERAL_TIMEOUT,
)
if conversation:
logger.info(
@@ -409,6 +605,7 @@ class SlackFactory:
conversation_id=conversation.conversation_id,
slack_conversation=conversation,
team_id=team_id,
v1_enabled=False,
)
elif SlackFactory.did_user_select_repo_from_form(message):
@@ -426,6 +623,7 @@ class SlackFactory:
send_summary_instruction=True,
conversation_id='',
team_id=team_id,
v1_enabled=False,
)
else:
@@ -443,4 +641,13 @@ class SlackFactory:
send_summary_instruction=True,
conversation_id='',
team_id=team_id,
v1_enabled=False,
)
# Type alias for all authenticated Slack view types that can start conversations
SlackViewType = (
SlackNewConversationView
| SlackNewConversationFromRepoFormView
| SlackUpdateExistingConversationView
)

View File

@@ -0,0 +1,53 @@
from storage.repository_store import RepositoryStore
from storage.stored_repository import StoredRepository
from storage.user_repo_map import UserRepositoryMap
from storage.user_repo_map_store import UserRepositoryMapStore
from openhands.core.config.openhands_config import OpenHandsConfig
from openhands.core.logger import openhands_logger as logger
from openhands.integrations.service_types import Repository
async def store_repositories_in_db(repos: list[Repository], user_id: str) -> None:
"""
Store repositories in DB and create user-repository mappings
Args:
repos: List of Repository objects to store
user_id: User ID associated with these repositories
"""
# Convert Repository objects to StoredRepository objects
# Convert Repository objects to UserRepositoryMap objects
stored_repos = []
user_repos = []
for repo in repos:
repo_id = f'{repo.git_provider.value}##{str(repo.id)}'
stored_repo = StoredRepository(
repo_name=repo.full_name,
repo_id=repo_id,
is_public=repo.is_public,
# Optional fields set to None by default
has_microagent=None,
has_setup_script=None,
)
stored_repos.append(stored_repo)
user_repo_map = UserRepositoryMap(user_id=user_id, repo_id=repo_id, admin=None)
user_repos.append(user_repo_map)
# Get config instance
config = OpenHandsConfig()
try:
# Store repositories in the repos table
repo_store = RepositoryStore.get_instance(config)
await repo_store.store_projects(stored_repos)
# Store user-repository mappings in the user-repos table
user_repo_store = UserRepositoryMapStore.get_instance(config)
await user_repo_store.store_user_repo_mappings(user_repos)
logger.info(f'Saved repos for user {user_id}')
except Exception:
logger.warning('Failed to save repos', exc_info=True)

View File

@@ -1,66 +1,93 @@
from uuid import UUID
import stripe
from server.auth.token_manager import TokenManager
from server.constants import STRIPE_API_KEY
from server.logger import logger
from storage.database import session_maker
from sqlalchemy import select
from storage.database import a_session_maker
from storage.org import Org
from storage.org_store import OrgStore
from storage.stripe_customer import StripeCustomer
stripe.api_key = STRIPE_API_KEY
async def find_customer_id_by_user_id(user_id: str) -> str | None:
# First search our own DB...
with session_maker() as session:
stripe_customer = (
session.query(StripeCustomer)
.filter(StripeCustomer.keycloak_user_id == user_id)
.first()
)
async def find_customer_id_by_org_id(org_id: UUID) -> str | None:
async with a_session_maker() as session:
stmt = select(StripeCustomer).where(StripeCustomer.org_id == org_id)
result = await session.execute(stmt)
stripe_customer = result.scalar_one_or_none()
if stripe_customer:
return stripe_customer.stripe_customer_id
# If that fails, fallback to stripe
search_result = await stripe.Customer.search_async(
query=f"metadata['user_id']:'{user_id}'",
query=f"metadata['org_id']:'{str(org_id)}'",
)
data = search_result.data
if not data:
logger.info('no_customer_for_user_id', extra={'user_id': user_id})
logger.info(
'no_customer_for_org_id',
extra={'org_id': str(org_id)},
)
return None
return data[0].id # type: ignore [attr-defined]
async def find_or_create_customer(user_id: str) -> str:
customer_id = await find_customer_id_by_user_id(user_id)
if customer_id:
return customer_id
logger.info('creating_customer', extra={'user_id': user_id})
async def find_customer_id_by_user_id(user_id: str) -> str | None:
# First search our own DB...
org = await OrgStore.get_current_org_from_keycloak_user_id(user_id)
if not org:
logger.warning(f'Org not found for user {user_id}')
return None
customer_id = await find_customer_id_by_org_id(org.id)
return customer_id
# Get the user info from keycloak
token_manager = TokenManager()
user_info = await token_manager.get_user_info_from_user_id(user_id) or {}
async def find_or_create_customer_by_user_id(user_id: str) -> dict | None:
# Get the current org for the user
org = await OrgStore.get_current_org_from_keycloak_user_id(user_id)
if not org:
logger.warning(f'Org not found for user {user_id}')
return None
customer_id = await find_customer_id_by_org_id(org.id)
if customer_id:
return {'customer_id': customer_id, 'org_id': str(org.id)}
logger.info(
'creating_customer',
extra={'user_id': user_id, 'org_id': str(org.id)},
)
# Create the customer in stripe
customer = await stripe.Customer.create_async(
email=str(user_info.get('email', '')),
metadata={'user_id': user_id},
email=org.contact_email,
metadata={'org_id': str(org.id)},
)
# Save the stripe customer in the local db
with session_maker() as session:
async with a_session_maker() as session:
session.add(
StripeCustomer(keycloak_user_id=user_id, stripe_customer_id=customer.id)
StripeCustomer(
keycloak_user_id=user_id,
org_id=org.id,
stripe_customer_id=customer.id,
)
)
session.commit()
await session.commit()
logger.info(
'created_customer',
extra={'user_id': user_id, 'stripe_customer_id': customer.id},
extra={
'user_id': user_id,
'org_id': str(org.id),
'stripe_customer_id': customer.id,
},
)
return customer.id
return {'customer_id': customer.id, 'org_id': str(org.id)}
async def has_payment_method(user_id: str) -> bool:
async def has_payment_method_by_user_id(user_id: str) -> bool:
customer_id = await find_customer_id_by_user_id(user_id)
if customer_id is None:
return False
@@ -71,3 +98,29 @@ async def has_payment_method(user_id: str) -> bool:
f'has_payment_method:{user_id}:{customer_id}:{bool(payment_methods.data)}'
)
return bool(payment_methods.data)
async def migrate_customer(user_id: str, org: Org):
async with a_session_maker() as session:
result = await session.execute(
select(StripeCustomer).where(StripeCustomer.keycloak_user_id == user_id)
)
stripe_customer = result.scalar_one_or_none()
if stripe_customer is None:
return
stripe_customer.org_id = org.id
customer = await stripe.Customer.modify_async(
id=stripe_customer.stripe_customer_id,
email=org.contact_email,
metadata={'user_id': '', 'org_id': str(org.id)},
)
logger.info(
'migrated_customer',
extra={
'user_id': user_id,
'org_id': str(org.id),
'stripe_customer_id': customer.id,
},
)
await session.commit()

View File

@@ -1,9 +1,17 @@
from dataclasses import dataclass
from enum import Enum
from typing import TYPE_CHECKING
from jinja2 import Environment
from pydantic import BaseModel
if TYPE_CHECKING:
from integrations.models import Message
from openhands.integrations.provider import PROVIDER_TOKEN_TYPE
from openhands.server.user_auth.user_auth import UserAuth
from openhands.storage.data_models.conversation_metadata import ConversationMetadata
class GitLabResourceType(Enum):
GROUP = 'group'
@@ -31,21 +39,41 @@ class SummaryExtractionTracker:
@dataclass
class ResolverViewInterface(SummaryExtractionTracker):
installation_id: int
# installation_id type varies by provider:
# - GitHub: int (GitHub App installation ID)
# - GitLab: str (webhook installation ID from our DB)
installation_id: int | str
user_info: UserData
issue_number: int
full_repo_name: str
is_public_repo: bool
raw_payload: dict
raw_payload: 'Message'
def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"Instructions passed when conversation is first initialized"
async def _get_instructions(self, jinja_env: Environment) -> tuple[str, str]:
"""Instructions passed when conversation is first initialized."""
raise NotImplementedError()
async def create_new_conversation(self, jinja_env: Environment, token: str):
"Create a new conversation"
async def initialize_new_conversation(self) -> 'ConversationMetadata':
"""Initialize a new conversation and return metadata.
For V1 conversations, creates a dummy ConversationMetadata.
For V0 conversations, initializes through the conversation store.
"""
raise NotImplementedError()
def get_callback_id(self) -> str:
"Unique callback id for subscribription made to EventStream for fetching agent summary"
async def create_new_conversation(
self,
jinja_env: Environment,
git_provider_tokens: 'PROVIDER_TOKEN_TYPE',
conversation_metadata: 'ConversationMetadata',
saas_user_auth: 'UserAuth',
) -> None:
"""Create a new conversation.
Args:
jinja_env: Jinja2 environment for template rendering
git_provider_tokens: Token mapping for git providers
conversation_metadata: Metadata for the conversation
saas_user_auth: User authentication for SaaS
"""
raise NotImplementedError()

View File

@@ -7,12 +7,8 @@ from typing import TYPE_CHECKING
from jinja2 import Environment, FileSystemLoader
from server.constants import WEB_HOST
from storage.repository_store import RepositoryStore
from storage.stored_repository import StoredRepository
from storage.user_repo_map import UserRepositoryMap
from storage.user_repo_map_store import UserRepositoryMapStore
from storage.org_store import OrgStore
from openhands.core.config.openhands_config import OpenHandsConfig
from openhands.core.logger import openhands_logger as logger
from openhands.core.schema.agent import AgentState
from openhands.events import Event, EventSource
@@ -20,6 +16,7 @@ from openhands.events.action import (
AgentFinishAction,
MessageAction,
)
from openhands.events.event_filter import EventFilter
from openhands.events.event_store_abc import EventStoreABC
from openhands.events.observation.agent import AgentStateChangedObservation
from openhands.integrations.service_types import Repository
@@ -35,7 +32,8 @@ if TYPE_CHECKING:
HOST = WEB_HOST
# ---- DO NOT REMOVE ----
HOST_URL = f'https://{HOST}'
IS_LOCAL_DEPLOYMENT = 'localhost' in HOST
HOST_URL = f'https://{HOST}' if not IS_LOCAL_DEPLOYMENT else f'http://{HOST}'
GITHUB_WEBHOOK_URL = f'{HOST_URL}/integration/github/events'
GITLAB_WEBHOOK_URL = f'{HOST_URL}/integration/gitlab/events'
conversation_prefix = 'conversations/{}'
@@ -46,6 +44,46 @@ ENABLE_PROACTIVE_CONVERSATION_STARTERS = (
os.getenv('ENABLE_PROACTIVE_CONVERSATION_STARTERS', 'false').lower() == 'true'
)
def get_session_expired_message(username: str | None = None) -> str:
"""Get a user-friendly session expired message.
Used by integrations to notify users when their Keycloak offline session
has expired.
Args:
username: Optional username to mention in the message. If provided,
the message will include @username prefix (used by Git providers
like GitHub, GitLab, Slack). If None, returns a generic message
(used by Jira, Jira DC, Linear).
Returns:
A formatted session expired message
"""
if username:
return f'@{username} your session has expired. Please login again at [OpenHands Cloud]({HOST_URL}) and try again.'
return f'Your session has expired. Please login again at [OpenHands Cloud]({HOST_URL}) and try again.'
def get_user_not_found_message(username: str | None = None) -> str:
"""Get a user-friendly message when a user hasn't created an OpenHands account.
Used by integrations to notify users when they try to use OpenHands features
but haven't logged into OpenHands Cloud yet (no Keycloak account exists).
Args:
username: Optional username to mention in the message. If provided,
the message will include @username prefix (used by Git providers
like GitHub, GitLab, Slack). If None, returns a generic message.
Returns:
A formatted user not found message
"""
if username:
return f"@{username} it looks like you haven't created an OpenHands account yet. Please sign up at [OpenHands Cloud]({HOST_URL}) and try again."
return f"It looks like you haven't created an OpenHands account yet. Please sign up at [OpenHands Cloud]({HOST_URL}) and try again."
# Toggle for solvability report feature
ENABLE_SOLVABILITY_ANALYSIS = (
os.getenv('ENABLE_SOLVABILITY_ANALYSIS', 'false').lower() == 'true'
@@ -56,8 +94,19 @@ ENABLE_V1_GITHUB_RESOLVER = (
os.getenv('ENABLE_V1_GITHUB_RESOLVER', 'false').lower() == 'true'
)
ENABLE_V1_SLACK_RESOLVER = (
os.getenv('ENABLE_V1_SLACK_RESOLVER', 'false').lower() == 'true'
)
OPENHANDS_RESOLVER_TEMPLATES_DIR = 'openhands/integrations/templates/resolver/'
# Toggle for V1 GitLab resolver feature
ENABLE_V1_GITLAB_RESOLVER = (
os.getenv('ENABLE_V1_GITLAB_RESOLVER', 'false').lower() == 'true'
)
OPENHANDS_RESOLVER_TEMPLATES_DIR = (
os.getenv('OPENHANDS_RESOLVER_TEMPLATES_DIR')
or 'openhands/integrations/templates/resolver/'
)
jinja_env = Environment(loader=FileSystemLoader(OPENHANDS_RESOLVER_TEMPLATES_DIR))
@@ -85,6 +134,26 @@ def get_summary_instruction():
return summary_instruction
async def get_user_v1_enabled_setting(user_id: str | None) -> bool:
"""Get the user's V1 conversation API setting.
Args:
user_id: The keycloak user ID
Returns:
True if V1 conversations are enabled for this user, False otherwise
"""
if not user_id:
return False
org = await OrgStore.get_current_org_from_keycloak_user_id(user_id)
if not org or org.v1_enabled is None:
return False
return org.v1_enabled
def has_exact_mention(text: str, mention: str) -> bool:
"""Check if the text contains an exact mention (not part of a larger word).
@@ -203,18 +272,35 @@ def get_summary_for_agent_state(
def get_final_agent_observation(
event_store: EventStoreABC,
) -> list[AgentStateChangedObservation]:
return event_store.get_matching_events(
source=EventSource.ENVIRONMENT,
event_types=(AgentStateChangedObservation,),
limit=1,
reverse=True,
events = list(
event_store.search_events(
filter=EventFilter(
source=EventSource.ENVIRONMENT,
include_types=(AgentStateChangedObservation,),
),
limit=1,
reverse=True,
)
)
result = [e for e in events if isinstance(e, AgentStateChangedObservation)]
assert len(result) == len(events)
return result
def get_last_user_msg(event_store: EventStoreABC) -> list[MessageAction]:
return event_store.get_matching_events(
source=EventSource.USER, event_types=(MessageAction,), limit=1, reverse='true'
events = list(
event_store.search_events(
filter=EventFilter(
source=EventSource.USER,
include_types=(MessageAction,),
),
limit=1,
reverse=True,
)
)
result = [e for e in events if isinstance(e, MessageAction)]
assert len(result) == len(events)
return result
def extract_summary_from_event_store(
@@ -226,18 +312,22 @@ def extract_summary_from_event_store(
conversation_link = CONVERSATION_URL.format(conversation_id)
summary_instruction = get_summary_instruction()
instruction_event: list[MessageAction] = event_store.get_matching_events(
query=json.dumps(summary_instruction),
source=EventSource.USER,
event_types=(MessageAction,),
limit=1,
reverse=True,
instruction_events = list(
event_store.search_events(
filter=EventFilter(
query=json.dumps(summary_instruction),
source=EventSource.USER,
include_types=(MessageAction,),
),
limit=1,
reverse=True,
)
)
final_agent_observation = get_final_agent_observation(event_store)
# Find summary instruction event ID
if len(instruction_event) == 0:
if not instruction_events:
logger.warning(
'no_instruction_event_found', extra={'conversation_id': conversation_id}
)
@@ -245,19 +335,19 @@ def extract_summary_from_event_store(
final_agent_observation, conversation_link
) # Agent did not receive summary instruction
event_id: int = instruction_event[0].id
agent_messages: list[MessageAction | AgentFinishAction] = (
event_store.get_matching_events(
start_id=event_id,
source=EventSource.AGENT,
event_types=(MessageAction, AgentFinishAction),
reverse=True,
summary_events = list(
event_store.search_events(
filter=EventFilter(
source=EventSource.AGENT,
include_types=(MessageAction, AgentFinishAction),
),
limit=1,
reverse=True,
start_id=instruction_events[0].id,
)
)
if len(agent_messages) == 0:
if not summary_events:
logger.warning(
'no_agent_messages_found', extra={'conversation_id': conversation_id}
)
@@ -265,10 +355,11 @@ def extract_summary_from_event_store(
final_agent_observation, conversation_link
) # Agent failed to generate summary
summary_event: MessageAction | AgentFinishAction = agent_messages[0]
summary_event = summary_events[0]
if isinstance(summary_event, MessageAction):
return summary_event.content
assert isinstance(summary_event, AgentFinishAction)
return summary_event.final_thought
@@ -321,106 +412,50 @@ def append_conversation_footer(message: str, conversation_id: str) -> str:
The message with the conversation footer appended
"""
conversation_link = CONVERSATION_URL.format(conversation_id)
footer = f'\n\n<sub>[View full conversation]({conversation_link})</sub>'
footer = f'\n\n[View full conversation]({conversation_link})'
return message + footer
async def store_repositories_in_db(repos: list[Repository], user_id: str) -> None:
"""
Store repositories in DB and create user-repository mappings
Args:
repos: List of Repository objects to store
user_id: User ID associated with these repositories
"""
# Convert Repository objects to StoredRepository objects
# Convert Repository objects to UserRepositoryMap objects
stored_repos = []
user_repos = []
for repo in repos:
repo_id = f'{repo.git_provider.value}##{str(repo.id)}'
stored_repo = StoredRepository(
repo_name=repo.full_name,
repo_id=repo_id,
is_public=repo.is_public,
# Optional fields set to None by default
has_microagent=None,
has_setup_script=None,
)
stored_repos.append(stored_repo)
user_repo_map = UserRepositoryMap(user_id=user_id, repo_id=repo_id, admin=None)
user_repos.append(user_repo_map)
# Get config instance
config = OpenHandsConfig()
try:
# Store repositories in the repos table
repo_store = RepositoryStore.get_instance(config)
repo_store.store_projects(stored_repos)
# Store user-repository mappings in the user-repos table
user_repo_store = UserRepositoryMapStore.get_instance(config)
user_repo_store.store_user_repo_mappings(user_repos)
logger.info(f'Saved repos for user {user_id}')
except Exception:
logger.warning('Failed to save repos', exc_info=True)
def infer_repo_from_message(user_msg: str) -> list[str]:
"""
Extract all repository names in the format 'owner/repo' from various Git provider URLs
and direct mentions in text. Supports GitHub, GitLab, and BitBucket.
Args:
user_msg: Input message that may contain repository references
Returns:
List of repository names in 'owner/repo' format, empty list if none found
"""
# Normalize the message by removing extra whitespace and newlines
normalized_msg = re.sub(r'\s+', ' ', user_msg.strip())
# Pattern to match Git URLs from GitHub, GitLab, and BitBucket
# Captures: protocol, domain, owner, repo (with optional .git extension)
git_url_pattern = r'https?://(?:github\.com|gitlab\.com|bitbucket\.org)/([a-zA-Z0-9_.-]+)/([a-zA-Z0-9_.-]+?)(?:\.git)?(?:[/?#].*?)?(?=\s|$|[^\w.-])'
# Pattern to match direct owner/repo mentions (e.g., "OpenHands/OpenHands")
# Must be surrounded by word boundaries or specific characters to avoid false positives
direct_pattern = (
r'(?:^|\s|[\[\(\'"])([a-zA-Z0-9_.-]+)/([a-zA-Z0-9_.-]+)(?=\s|$|[\]\)\'",.])'
git_url_pattern = (
r'https?://(?:github\.com|gitlab\.com|bitbucket\.org)/'
r'([a-zA-Z0-9_.-]+)/([a-zA-Z0-9_.-]+?)(?:\.git)?'
r'(?:[/?#].*?)?(?=\s|$|[^\w.-])'
)
matches = []
# UPDATED: allow {{ owner/repo }} in addition to existing boundaries
direct_pattern = (
r'(?:^|\s|{{|[\[\(\'":`])' # left boundary
r'([a-zA-Z0-9_.-]+)/([a-zA-Z0-9_.-]+)'
r'(?=\s|$|}}|[\]\)\'",.:`])' # right boundary
)
# First, find all Git URLs (highest priority)
git_matches = re.findall(git_url_pattern, normalized_msg)
for owner, repo in git_matches:
# Remove .git extension if present
matches: list[str] = []
# Git URLs first (highest priority)
for owner, repo in re.findall(git_url_pattern, normalized_msg):
repo = re.sub(r'\.git$', '', repo)
matches.append(f'{owner}/{repo}')
# Second, find all direct owner/repo mentions
direct_matches = re.findall(direct_pattern, normalized_msg)
for owner, repo in direct_matches:
# Direct mentions
for owner, repo in re.findall(direct_pattern, normalized_msg):
full_match = f'{owner}/{repo}'
# Skip if it looks like a version number, date, or file path
if (
re.match(r'^\d+\.\d+/\d+\.\d+$', full_match) # version numbers
or re.match(r'^\d{1,2}/\d{1,2}$', full_match) # dates
or re.match(r'^[A-Z]/[A-Z]$', full_match) # single letters
or repo.endswith('.txt')
or repo.endswith('.md') # file extensions
or repo.endswith('.py')
or repo.endswith('.js')
or '.' in repo
and len(repo.split('.')) > 2
): # complex file paths
re.match(r'^\d+\.\d+/\d+\.\d+$', full_match)
or re.match(r'^\d{1,2}/\d{1,2}$', full_match)
or re.match(r'^[A-Z]/[A-Z]$', full_match)
or repo.endswith(('.txt', '.md', '.py', '.js'))
or ('.' in repo and len(repo.split('.')) > 2)
):
continue
# Avoid duplicates from Git URLs already found
if full_match not in matches:
matches.append(full_match)

View File

@@ -1,3 +1,8 @@
import logging
from typing import Callable, Coroutine
from uuid import UUID
from integrations.utils import CONVERSATION_URL
from pydantic import SecretStr
from server.auth.saas_user_auth import SaasUserAuth
from server.auth.token_manager import TokenManager
@@ -6,6 +11,78 @@ from openhands.core.logger import openhands_logger as logger
from openhands.server.user_auth.user_auth import UserAuth
def is_budget_exceeded_error(error_message: str) -> bool:
"""Check if an error message indicates a budget exceeded condition.
This is used to downgrade error logs to info logs for budget exceeded errors
since they are expected cost control behavior rather than unexpected errors.
"""
lower_message = error_message.lower()
return 'budget' in lower_message and 'exceeded' in lower_message
BUDGET_EXCEEDED_USER_MESSAGE = 'LLM budget has been exceeded, please re-fill.'
async def handle_callback_error(
error: Exception,
conversation_id: UUID,
service_name: str,
service_logger: logging.Logger,
can_post_error: bool,
post_error_func: Callable[[str], Coroutine],
) -> None:
"""Handle callback processing errors with appropriate logging and user messages.
This centralizes the error handling logic for V1 callback processors to:
- Log budget exceeded errors at INFO level (expected cost control behavior)
- Log other errors at EXCEPTION level
- Post user-friendly error messages to the integration platform
Args:
error: The exception that occurred
conversation_id: The conversation ID for logging and linking
service_name: The service name for log messages (e.g., "GitHub", "GitLab", "Slack")
service_logger: The logger instance to use for logging
can_post_error: Whether the prerequisites are met to post an error message
post_error_func: Async function to post the error message to the platform
"""
error_str = str(error)
budget_exceeded = is_budget_exceeded_error(error_str)
# Log appropriately based on error type
if budget_exceeded:
service_logger.info(
'[%s V1] Budget exceeded for conversation %s: %s',
service_name,
conversation_id,
error,
)
else:
service_logger.exception(
'[%s V1] Error processing callback: %s', service_name, error
)
# Try to post error message to the platform
if can_post_error:
try:
error_detail = (
BUDGET_EXCEEDED_USER_MESSAGE if budget_exceeded else error_str
)
await post_error_func(
f'OpenHands encountered an error: **{error_detail}**\n\n'
f'[See the conversation]({CONVERSATION_URL.format(conversation_id)}) '
'for more information.'
)
except Exception as post_error:
service_logger.warning(
'[%s V1] Failed to post error message to %s: %s',
service_name,
service_name,
post_error,
)
async def get_saas_user_auth(
keycloak_user_id: str, token_manager: TokenManager
) -> UserAuth:

View File

@@ -1,10 +1,15 @@
import logging
import os
from logging.config import fileConfig
from alembic import context
from google.cloud.sql.connector import Connector
from sqlalchemy import create_engine
from storage.base import Base
# Suppress alembic.runtime.plugins INFO logs during import to prevent non-JSON logs in production
# These plugin setup messages would otherwise appear before logging is configured
logging.getLogger('alembic.runtime.plugins').setLevel(logging.WARNING)
from alembic import context # noqa: E402
from google.cloud.sql.connector import Connector # noqa: E402
from sqlalchemy import create_engine # noqa: E402
from storage.base import Base # noqa: E402
target_metadata = Base.metadata

View File

@@ -20,6 +20,8 @@ down_revision = '059'
branch_labels = None
depends_on = None
# TODO: decide whether to modify this for orgs or users
def upgrade():
"""
@@ -28,8 +30,10 @@ def upgrade():
This replaces the functionality of the removed admin maintenance endpoint.
"""
# Import here to avoid circular imports
from server.constants import CURRENT_USER_SETTINGS_VERSION
# Hardcoded value to prevent migration failures when constant is removed from codebase
# This migration has already run in production, so we use the value that was current at the time
CURRENT_USER_SETTINGS_VERSION = 4
# Create a connection and bind it to a session
connection = op.get_bind()

View File

@@ -0,0 +1,30 @@
"""add v1 column to slack conversation table
Revision ID: 086
Revises: 085
Create Date: 2025-12-02 15:30:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
# revision identifiers, used by Alembic.
revision: str = '086'
down_revision: Union[str, None] = '085'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# Add v1 column
op.add_column(
'slack_conversation', sa.Column('v1_enabled', sa.Boolean(), nullable=True)
)
def downgrade() -> None:
# Drop v1 column
op.drop_column('slack_conversation', 'v1_enabled')

View File

@@ -0,0 +1,61 @@
"""bump condenser defaults: max_size 120->240
Revision ID: 087
Revises: 086
Create Date: 2026-01-05
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
from sqlalchemy.sql import column, table
# revision identifiers, used by Alembic.
revision: str = '087'
down_revision: Union[str, None] = '086'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
"""Upgrade schema.
Update existing users with condenser_max_size=120 or NULL to 240.
This covers both users who had the old default (120) explicitly set
and users who had NULL (which defaulted to 120 in the application code).
The SDK default for keep_first will be used automatically.
"""
user_settings_table = table(
'user_settings',
column('condenser_max_size', sa.Integer),
)
# Update users with explicit 120 value
op.execute(
user_settings_table.update()
.where(user_settings_table.c.condenser_max_size == 120)
.values(condenser_max_size=240)
)
# Update users with NULL value (which defaulted to 120 in application code)
op.execute(
user_settings_table.update()
.where(user_settings_table.c.condenser_max_size.is_(None))
.values(condenser_max_size=240)
)
def downgrade() -> None:
"""Downgrade schema.
Note: This sets all 240 values back to NULL (not 120) since we can't
distinguish between users who had 120 vs NULL before the upgrade.
"""
user_settings_table = table(
'user_settings', column('condenser_max_size', sa.Integer)
)
op.execute(
user_settings_table.update()
.where(user_settings_table.c.condenser_max_size == 240)
.values(condenser_max_size=None)
)

View File

@@ -0,0 +1,54 @@
"""create blocked_email_domains table
Revision ID: 088
Revises: 087
Create Date: 2025-01-27 00:00:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
# revision identifiers, used by Alembic.
revision: str = '088'
down_revision: Union[str, None] = '087'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
"""Create blocked_email_domains table for storing blocked email domain patterns."""
op.create_table(
'blocked_email_domains',
sa.Column('id', sa.Integer(), sa.Identity(), nullable=False, primary_key=True),
sa.Column('domain', sa.String(), nullable=False),
sa.Column(
'created_at',
sa.DateTime(timezone=True),
nullable=False,
server_default=sa.text('CURRENT_TIMESTAMP'),
),
sa.Column(
'updated_at',
sa.DateTime(timezone=True),
nullable=False,
server_default=sa.text('CURRENT_TIMESTAMP'),
),
sa.PrimaryKeyConstraint('id'),
)
# Create unique index on domain column
op.create_index(
'ix_blocked_email_domains_domain',
'blocked_email_domains',
['domain'],
unique=True,
)
def downgrade() -> None:
"""Drop blocked_email_domains table."""
op.drop_index('ix_blocked_email_domains_domain', table_name='blocked_email_domains')
op.drop_table('blocked_email_domains')

View File

@@ -0,0 +1,272 @@
"""create org tables from pgerd schema
Revision ID: 089
Revises: 088
Create Date: 2025-01-07 00:00:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
from sqlalchemy.dialects import postgresql
# revision identifiers, used by Alembic.
revision: str = '089'
down_revision: Union[str, None] = '088'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# Remove current settings table
op.execute('DROP TABLE IF EXISTS settings')
# Add already_migrated column to user_settings table
op.add_column(
'user_settings',
sa.Column(
'already_migrated',
sa.Boolean,
nullable=True,
server_default=sa.text('false'),
),
)
# Create role table
op.create_table(
'role',
sa.Column('id', sa.Integer, sa.Identity(), primary_key=True),
sa.Column('name', sa.String, nullable=False),
sa.Column('rank', sa.Integer, nullable=False),
sa.UniqueConstraint('name', name='role_name_unique'),
)
# 1. Create default roles
op.execute(
sa.text("""
INSERT INTO role (name, rank) VALUES ('owner', 10), ('admin', 20), ('user', 1000)
ON CONFLICT (name) DO NOTHING;
""")
)
# Create org table with settings fields
op.create_table(
'org',
sa.Column(
'id',
postgresql.UUID(as_uuid=True),
primary_key=True,
),
sa.Column('name', sa.String, nullable=False),
sa.Column('contact_name', sa.String, nullable=True),
sa.Column('contact_email', sa.String, nullable=True),
sa.Column('conversation_expiration', sa.Integer, nullable=True),
# Settings fields moved to org table
sa.Column('agent', sa.String, nullable=True),
sa.Column('default_max_iterations', sa.Integer, nullable=True),
sa.Column('security_analyzer', sa.String, nullable=True),
sa.Column(
'confirmation_mode',
sa.Boolean,
nullable=True,
server_default=sa.text('false'),
),
sa.Column('default_llm_model', sa.String, nullable=True),
sa.Column('default_llm_base_url', sa.String, nullable=True),
sa.Column('remote_runtime_resource_factor', sa.Integer, nullable=True),
sa.Column(
'enable_default_condenser',
sa.Boolean,
nullable=False,
server_default=sa.text('true'),
),
sa.Column('billing_margin', sa.Float, nullable=True),
sa.Column(
'enable_proactive_conversation_starters',
sa.Boolean,
nullable=False,
server_default=sa.text('true'),
),
sa.Column('sandbox_base_container_image', sa.String, nullable=True),
sa.Column('sandbox_runtime_container_image', sa.String, nullable=True),
sa.Column(
'org_version', sa.Integer, nullable=False, server_default=sa.text('0')
),
sa.Column('mcp_config', sa.JSON, nullable=True),
sa.Column('_search_api_key', sa.String, nullable=True),
sa.Column('_sandbox_api_key', sa.String, nullable=True),
sa.Column('max_budget_per_task', sa.Float, nullable=True),
sa.Column(
'enable_solvability_analysis',
sa.Boolean,
nullable=True,
server_default=sa.text('false'),
),
sa.Column('v1_enabled', sa.Boolean, nullable=True),
sa.Column('condenser_max_size', sa.Integer, nullable=True),
sa.UniqueConstraint('name', name='org_name_unique'),
)
# Create user table with user-specific settings fields
op.create_table(
'user',
sa.Column(
'id',
postgresql.UUID(as_uuid=True),
primary_key=True,
),
sa.Column('current_org_id', postgresql.UUID(as_uuid=True), nullable=False),
sa.Column('role_id', sa.Integer, nullable=True),
sa.Column('accepted_tos', sa.DateTime, nullable=True),
sa.Column(
'enable_sound_notifications',
sa.Boolean,
nullable=True,
server_default=sa.text('false'),
),
sa.Column('language', sa.String, nullable=True),
sa.Column('user_consents_to_analytics', sa.Boolean, nullable=True),
sa.Column('email', sa.String, nullable=True),
sa.Column('email_verified', sa.Boolean, nullable=True),
sa.ForeignKeyConstraint(
['current_org_id'], ['org.id'], name='current_org_fkey'
),
sa.ForeignKeyConstraint(['role_id'], ['role.id'], name='user_role_fkey'),
)
# Create org_member table (junction table for many-to-many relationship)
op.create_table(
'org_member',
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=False),
sa.Column('user_id', postgresql.UUID(as_uuid=True), nullable=False),
sa.Column('role_id', sa.Integer, nullable=False),
sa.Column('_llm_api_key', sa.String, nullable=False),
sa.Column('max_iterations', sa.Integer, nullable=True),
sa.Column('llm_model', sa.String, nullable=True),
sa.Column('_llm_api_key_for_byor', sa.String, nullable=True),
sa.Column('llm_base_url', sa.String, nullable=True),
sa.Column('status', sa.String, nullable=True),
sa.ForeignKeyConstraint(['org_id'], ['org.id'], name='om_org_fkey'),
sa.ForeignKeyConstraint(['user_id'], ['user.id'], name='om_user_fkey'),
sa.ForeignKeyConstraint(['role_id'], ['role.id'], name='om_role_fkey'),
sa.PrimaryKeyConstraint('org_id', 'user_id'),
)
# Add org_id column to existing tables
# billing_sessions
op.add_column(
'billing_sessions',
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True),
)
op.create_foreign_key(
'billing_sessions_org_fkey', 'billing_sessions', 'org', ['org_id'], ['id']
)
# Create conversation_metadata_saas table
op.create_table(
'conversation_metadata_saas',
sa.Column('conversation_id', sa.String(), nullable=False),
sa.Column('user_id', postgresql.UUID(as_uuid=True), nullable=False),
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=False),
sa.ForeignKeyConstraint(
['user_id'], ['user.id'], name='conversation_metadata_saas_user_fkey'
),
sa.ForeignKeyConstraint(
['org_id'], ['org.id'], name='conversation_metadata_saas_org_fkey'
),
sa.PrimaryKeyConstraint('conversation_id'),
)
# custom_secrets
op.add_column(
'custom_secrets',
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True),
)
op.create_foreign_key(
'custom_secrets_org_fkey', 'custom_secrets', 'org', ['org_id'], ['id']
)
# api_keys
op.add_column(
'api_keys', sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True)
)
op.create_foreign_key('api_keys_org_fkey', 'api_keys', 'org', ['org_id'], ['id'])
# slack_conversation
op.add_column(
'slack_conversation',
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True),
)
op.create_foreign_key(
'slack_conversation_org_fkey', 'slack_conversation', 'org', ['org_id'], ['id']
)
# slack_users
op.add_column(
'slack_users', sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True)
)
op.create_foreign_key(
'slack_users_org_fkey', 'slack_users', 'org', ['org_id'], ['id']
)
# stripe_customers
op.alter_column(
'stripe_customers',
'keycloak_user_id',
existing_type=sa.String(),
nullable=True,
)
op.add_column(
'stripe_customers',
sa.Column('org_id', postgresql.UUID(as_uuid=True), nullable=True),
)
op.create_foreign_key(
'stripe_customers_org_fkey', 'stripe_customers', 'org', ['org_id'], ['id']
)
def downgrade() -> None:
# Drop already_migrated column from user_settings table
op.drop_column('user_settings', 'already_migrated')
# Drop foreign keys and columns added to existing tables
op.drop_constraint(
'stripe_customers_org_fkey', 'stripe_customers', type_='foreignkey'
)
op.drop_column('stripe_customers', 'org_id')
op.alter_column(
'stripe_customers',
'keycloak_user_id',
existing_type=sa.String(),
nullable=False,
)
op.drop_constraint('slack_users_org_fkey', 'slack_users', type_='foreignkey')
op.drop_column('slack_users', 'org_id')
op.drop_constraint(
'slack_conversation_org_fkey', 'slack_conversation', type_='foreignkey'
)
op.drop_column('slack_conversation', 'org_id')
op.drop_constraint('api_keys_org_fkey', 'api_keys', type_='foreignkey')
op.drop_column('api_keys', 'org_id')
op.drop_constraint('custom_secrets_org_fkey', 'custom_secrets', type_='foreignkey')
op.drop_column('custom_secrets', 'org_id')
# Drop conversation_metadata_saas table
op.drop_table('conversation_metadata_saas')
op.drop_constraint(
'billing_sessions_org_fkey', 'billing_sessions', type_='foreignkey'
)
op.drop_column('billing_sessions', 'org_id')
# Drop tables in reverse order due to foreign key constraints
op.drop_table('org_member')
op.drop_table('user')
op.drop_table('org')
op.drop_table('role')

View File

@@ -0,0 +1,28 @@
"""Add git_user_name and git_user_email columns to user table.
Revision ID: 090
Revises: 089
Create Date: 2025-01-22
"""
import sqlalchemy as sa
from alembic import op
revision = '090'
down_revision = '089'
def upgrade() -> None:
op.add_column(
'user',
sa.Column('git_user_name', sa.String, nullable=True),
)
op.add_column(
'user',
sa.Column('git_user_email', sa.String, nullable=True),
)
def downgrade() -> None:
op.drop_column('user', 'git_user_email')
op.drop_column('user', 'git_user_name')

Some files were not shown because too many files have changed in this diff Show More