mirror of
https://github.com/OpenHands/OpenHands.git
synced 2025-12-26 05:48:36 +08:00
Support microagents in CLI and Headless (#5971)
Co-authored-by: Engel Nyst <enyst@users.noreply.github.com>
This commit is contained in:
Robert Brennan
GitHub
parent
9dd5463e06
commit
8983d719bd
@ -1,15 +1,12 @@
|
||||
import asyncio
|
||||
import logging
|
||||
import sys
|
||||
from typing import Type
|
||||
from uuid import uuid4
|
||||
|
||||
from termcolor import colored
|
||||
|
||||
import openhands.agenthub # noqa F401 (we import this to get the agents registered)
|
||||
from openhands import __version__
|
||||
from openhands.controller import AgentController
|
||||
from openhands.controller.agent import Agent
|
||||
from openhands.core.config import (
|
||||
AppConfig,
|
||||
get_parser,
|
||||
@ -18,7 +15,8 @@ from openhands.core.config import (
|
||||
from openhands.core.logger import openhands_logger as logger
|
||||
from openhands.core.loop import run_agent_until_done
|
||||
from openhands.core.schema import AgentState
|
||||
from openhands.events import EventSource, EventStream, EventStreamSubscriber
|
||||
from openhands.core.setup import create_agent, create_controller, create_runtime
|
||||
from openhands.events import EventSource, EventStreamSubscriber
|
||||
from openhands.events.action import (
|
||||
Action,
|
||||
ActionConfirmationStatus,
|
||||
@ -34,11 +32,6 @@ from openhands.events.observation import (
|
||||
FileEditObservation,
|
||||
NullObservation,
|
||||
)
|
||||
from openhands.llm.llm import LLM
|
||||
from openhands.runtime import get_runtime_cls
|
||||
from openhands.runtime.base import Runtime
|
||||
from openhands.security import SecurityAnalyzer, options
|
||||
from openhands.storage import get_file_store
|
||||
|
||||
|
||||
def display_message(message: str):
|
||||
@ -114,39 +107,12 @@ async def main(loop):
|
||||
config = load_app_config(config_file=args.config_file)
|
||||
sid = str(uuid4())
|
||||
|
||||
agent_cls: Type[Agent] = Agent.get_cls(config.default_agent)
|
||||
agent_config = config.get_agent_config(config.default_agent)
|
||||
llm_config = config.get_llm_config_from_agent(config.default_agent)
|
||||
agent = agent_cls(
|
||||
llm=LLM(config=llm_config),
|
||||
config=agent_config,
|
||||
)
|
||||
runtime = create_runtime(config, sid=sid, headless_mode=True)
|
||||
await runtime.connect()
|
||||
agent = create_agent(runtime, config)
|
||||
controller, _ = create_controller(agent, runtime, config)
|
||||
|
||||
file_store = get_file_store(config.file_store, config.file_store_path)
|
||||
event_stream = EventStream(sid, file_store)
|
||||
|
||||
runtime_cls = get_runtime_cls(config.runtime)
|
||||
runtime: Runtime = runtime_cls( # noqa: F841
|
||||
config=config,
|
||||
event_stream=event_stream,
|
||||
sid=sid,
|
||||
plugins=agent_cls.sandbox_plugins,
|
||||
headless_mode=True,
|
||||
)
|
||||
|
||||
if config.security.security_analyzer:
|
||||
options.SecurityAnalyzers.get(
|
||||
config.security.security_analyzer, SecurityAnalyzer
|
||||
)(event_stream)
|
||||
|
||||
controller = AgentController(
|
||||
agent=agent,
|
||||
max_iterations=config.max_iterations,
|
||||
max_budget_per_task=config.max_budget_per_task,
|
||||
agent_to_llm_config=config.get_agent_to_llm_config_map(),
|
||||
event_stream=event_stream,
|
||||
confirmation_mode=config.security.confirmation_mode,
|
||||
)
|
||||
event_stream = runtime.event_stream
|
||||
|
||||
async def prompt_for_next_task():
|
||||
# Run input() in a thread pool to avoid blocking the event loop
|
||||
|
||||
@ -1,13 +1,10 @@
|
||||
import asyncio
|
||||
import hashlib
|
||||
import json
|
||||
import os
|
||||
import sys
|
||||
import uuid
|
||||
from typing import Callable, Protocol, Type
|
||||
from typing import Callable, Protocol
|
||||
|
||||
import openhands.agenthub # noqa F401 (we import this to get the agents registered)
|
||||
from openhands.controller import AgentController
|
||||
from openhands.controller.agent import Agent
|
||||
from openhands.controller.state.state import State
|
||||
from openhands.core.config import (
|
||||
@ -19,16 +16,19 @@ from openhands.core.config import (
|
||||
from openhands.core.logger import openhands_logger as logger
|
||||
from openhands.core.loop import run_agent_until_done
|
||||
from openhands.core.schema import AgentState
|
||||
from openhands.events import EventSource, EventStream, EventStreamSubscriber
|
||||
from openhands.core.setup import (
|
||||
create_agent,
|
||||
create_controller,
|
||||
create_runtime,
|
||||
generate_sid,
|
||||
)
|
||||
from openhands.events import EventSource, EventStreamSubscriber
|
||||
from openhands.events.action import MessageAction
|
||||
from openhands.events.action.action import Action
|
||||
from openhands.events.event import Event
|
||||
from openhands.events.observation import AgentStateChangedObservation
|
||||
from openhands.events.serialization.event import event_to_trajectory
|
||||
from openhands.llm.llm import LLM
|
||||
from openhands.runtime import get_runtime_cls
|
||||
from openhands.runtime.base import Runtime
|
||||
from openhands.storage import get_file_store
|
||||
|
||||
|
||||
class FakeUserResponseFunc(Protocol):
|
||||
@ -51,45 +51,6 @@ def read_task_from_stdin() -> str:
|
||||
return sys.stdin.read()
|
||||
|
||||
|
||||
def create_runtime(
|
||||
config: AppConfig,
|
||||
sid: str | None = None,
|
||||
headless_mode: bool = True,
|
||||
) -> Runtime:
|
||||
"""Create a runtime for the agent to run on.
|
||||
|
||||
config: The app config.
|
||||
sid: (optional) The session id. IMPORTANT: please don't set this unless you know what you're doing.
|
||||
Set it to incompatible value will cause unexpected behavior on RemoteRuntime.
|
||||
headless_mode: Whether the agent is run in headless mode. `create_runtime` is typically called within evaluation scripts,
|
||||
where we don't want to have the VSCode UI open, so it defaults to True.
|
||||
"""
|
||||
# if sid is provided on the command line, use it as the name of the event stream
|
||||
# otherwise generate it on the basis of the configured jwt_secret
|
||||
# we can do this better, this is just so that the sid is retrieved when we want to restore the session
|
||||
session_id = sid or generate_sid(config)
|
||||
|
||||
# set up the event stream
|
||||
file_store = get_file_store(config.file_store, config.file_store_path)
|
||||
event_stream = EventStream(session_id, file_store)
|
||||
|
||||
# agent class
|
||||
agent_cls = openhands.agenthub.Agent.get_cls(config.default_agent)
|
||||
|
||||
# runtime and tools
|
||||
runtime_cls = get_runtime_cls(config.runtime)
|
||||
logger.debug(f'Initializing runtime: {runtime_cls.__name__}')
|
||||
runtime: Runtime = runtime_cls(
|
||||
config=config,
|
||||
event_stream=event_stream,
|
||||
sid=session_id,
|
||||
plugins=agent_cls.sandbox_plugins,
|
||||
headless_mode=headless_mode,
|
||||
)
|
||||
|
||||
return runtime
|
||||
|
||||
|
||||
async def run_controller(
|
||||
config: AppConfig,
|
||||
initial_user_action: Action,
|
||||
@ -115,17 +76,6 @@ async def run_controller(
|
||||
(could be None) and returns a fake user response.
|
||||
headless_mode: Whether the agent is run in headless mode.
|
||||
"""
|
||||
# Create the agent
|
||||
if agent is None:
|
||||
agent_cls: Type[Agent] = Agent.get_cls(config.default_agent)
|
||||
agent_config = config.get_agent_config(config.default_agent)
|
||||
llm_config = config.get_llm_config_from_agent(config.default_agent)
|
||||
agent = agent_cls(
|
||||
llm=LLM(config=llm_config),
|
||||
config=agent_config,
|
||||
)
|
||||
|
||||
# make sure the session id is set
|
||||
sid = sid or generate_sid(config)
|
||||
|
||||
if runtime is None:
|
||||
@ -134,28 +84,10 @@ async def run_controller(
|
||||
|
||||
event_stream = runtime.event_stream
|
||||
|
||||
# restore cli session if available
|
||||
initial_state = None
|
||||
try:
|
||||
logger.debug(
|
||||
f'Trying to restore agent state from cli session {event_stream.sid} if available'
|
||||
)
|
||||
initial_state = State.restore_from_session(
|
||||
event_stream.sid, event_stream.file_store
|
||||
)
|
||||
except Exception as e:
|
||||
logger.debug(f'Cannot restore agent state: {e}')
|
||||
if agent is None:
|
||||
agent = create_agent(runtime, config)
|
||||
|
||||
# init controller with this initial state
|
||||
controller = AgentController(
|
||||
agent=agent,
|
||||
max_iterations=config.max_iterations,
|
||||
max_budget_per_task=config.max_budget_per_task,
|
||||
agent_to_llm_config=config.get_agent_to_llm_config_map(),
|
||||
event_stream=event_stream,
|
||||
initial_state=initial_state,
|
||||
headless_mode=headless_mode,
|
||||
)
|
||||
controller, initial_state = create_controller(agent, runtime, config)
|
||||
|
||||
assert isinstance(
|
||||
initial_user_action, Action
|
||||
@ -234,15 +166,6 @@ async def run_controller(
|
||||
return state
|
||||
|
||||
|
||||
def generate_sid(config: AppConfig, session_name: str | None = None) -> str:
|
||||
"""Generate a session id based on the session name and the jwt secret."""
|
||||
session_name = session_name or str(uuid.uuid4())
|
||||
jwt_secret = config.jwt_secret
|
||||
|
||||
hash_str = hashlib.sha256(f'{session_name}{jwt_secret}'.encode('utf-8')).hexdigest()
|
||||
return f'{session_name}-{hash_str[:16]}'
|
||||
|
||||
|
||||
def auto_continue_response(
|
||||
state: State,
|
||||
encapsulate_solution: bool = False,
|
||||
|
||||
114
openhands/core/setup.py
Normal file
114
openhands/core/setup.py
Normal file
@ -0,0 +1,114 @@
|
||||
import hashlib
|
||||
import uuid
|
||||
from typing import Tuple, Type
|
||||
|
||||
import openhands.agenthub # noqa F401 (we import this to get the agents registered)
|
||||
from openhands.controller import AgentController
|
||||
from openhands.controller.agent import Agent
|
||||
from openhands.controller.state.state import State
|
||||
from openhands.core.config import (
|
||||
AppConfig,
|
||||
)
|
||||
from openhands.core.logger import openhands_logger as logger
|
||||
from openhands.events import EventStream
|
||||
from openhands.llm.llm import LLM
|
||||
from openhands.runtime import get_runtime_cls
|
||||
from openhands.runtime.base import Runtime
|
||||
from openhands.security import SecurityAnalyzer, options
|
||||
from openhands.storage import get_file_store
|
||||
|
||||
|
||||
def create_runtime(
|
||||
config: AppConfig,
|
||||
sid: str | None = None,
|
||||
headless_mode: bool = True,
|
||||
) -> Runtime:
|
||||
"""Create a runtime for the agent to run on.
|
||||
|
||||
config: The app config.
|
||||
sid: (optional) The session id. IMPORTANT: please don't set this unless you know what you're doing.
|
||||
Set it to incompatible value will cause unexpected behavior on RemoteRuntime.
|
||||
headless_mode: Whether the agent is run in headless mode. `create_runtime` is typically called within evaluation scripts,
|
||||
where we don't want to have the VSCode UI open, so it defaults to True.
|
||||
"""
|
||||
# if sid is provided on the command line, use it as the name of the event stream
|
||||
# otherwise generate it on the basis of the configured jwt_secret
|
||||
# we can do this better, this is just so that the sid is retrieved when we want to restore the session
|
||||
session_id = sid or generate_sid(config)
|
||||
|
||||
# set up the event stream
|
||||
file_store = get_file_store(config.file_store, config.file_store_path)
|
||||
event_stream = EventStream(session_id, file_store)
|
||||
|
||||
# agent class
|
||||
agent_cls = openhands.agenthub.Agent.get_cls(config.default_agent)
|
||||
|
||||
# runtime and tools
|
||||
runtime_cls = get_runtime_cls(config.runtime)
|
||||
logger.debug(f'Initializing runtime: {runtime_cls.__name__}')
|
||||
runtime: Runtime = runtime_cls(
|
||||
config=config,
|
||||
event_stream=event_stream,
|
||||
sid=session_id,
|
||||
plugins=agent_cls.sandbox_plugins,
|
||||
headless_mode=headless_mode,
|
||||
)
|
||||
|
||||
return runtime
|
||||
|
||||
|
||||
def create_agent(runtime: Runtime, config: AppConfig) -> Agent:
|
||||
agent_cls: Type[Agent] = Agent.get_cls(config.default_agent)
|
||||
agent_config = config.get_agent_config(config.default_agent)
|
||||
llm_config = config.get_llm_config_from_agent(config.default_agent)
|
||||
agent = agent_cls(
|
||||
llm=LLM(config=llm_config),
|
||||
config=agent_config,
|
||||
)
|
||||
if agent.prompt_manager:
|
||||
microagents = runtime.get_custom_microagents(None)
|
||||
agent.prompt_manager.load_microagent_files(microagents)
|
||||
|
||||
if config.security.security_analyzer:
|
||||
options.SecurityAnalyzers.get(
|
||||
config.security.security_analyzer, SecurityAnalyzer
|
||||
)(runtime.event_stream)
|
||||
|
||||
return agent
|
||||
|
||||
|
||||
def create_controller(
|
||||
agent: Agent, runtime: Runtime, config: AppConfig, headless_mode: bool = True
|
||||
) -> Tuple[AgentController, State | None]:
|
||||
event_stream = runtime.event_stream
|
||||
initial_state = None
|
||||
try:
|
||||
logger.debug(
|
||||
f'Trying to restore agent state from session {event_stream.sid} if available'
|
||||
)
|
||||
initial_state = State.restore_from_session(
|
||||
event_stream.sid, event_stream.file_store
|
||||
)
|
||||
except Exception as e:
|
||||
logger.debug(f'Cannot restore agent state: {e}')
|
||||
|
||||
controller = AgentController(
|
||||
agent=agent,
|
||||
max_iterations=config.max_iterations,
|
||||
max_budget_per_task=config.max_budget_per_task,
|
||||
agent_to_llm_config=config.get_agent_to_llm_config_map(),
|
||||
event_stream=event_stream,
|
||||
initial_state=initial_state,
|
||||
headless_mode=headless_mode,
|
||||
confirmation_mode=config.security.confirmation_mode,
|
||||
)
|
||||
return (controller, initial_state)
|
||||
|
||||
|
||||
def generate_sid(config: AppConfig, session_name: str | None = None) -> str:
|
||||
"""Generate a session id based on the session name and the jwt secret."""
|
||||
session_name = session_name or str(uuid.uuid4())
|
||||
jwt_secret = config.jwt_secret
|
||||
|
||||
hash_str = hashlib.sha256(f'{session_name}{jwt_secret}'.encode('utf-8')).hexdigest()
|
||||
return f'{session_name}-{hash_str[:16]}'
|
||||
Loading…
x
Reference in New Issue
Block a user