59 lines
1.8 KiB
Python
59 lines
1.8 KiB
Python
from pathlib import Path
|
|
|
|
from langchain.agents import create_agent
|
|
from langchain.agents.middleware import ModelRequest, ModelResponse, dynamic_prompt, wrap_model_call
|
|
|
|
from src import config as sys_config
|
|
from src.agents.common.base import BaseAgent
|
|
from src.agents.common.models import load_chat_model
|
|
from src.agents.common.tools import get_buildin_tools
|
|
from src.utils import logger
|
|
|
|
model = load_chat_model("siliconflow/Qwen/Qwen3-235B-A22B-Instruct-2507")
|
|
|
|
|
|
@dynamic_prompt
|
|
def context_aware_prompt(request: ModelRequest) -> str:
|
|
runtime = request.runtime
|
|
return runtime.context.system_prompt
|
|
|
|
|
|
@wrap_model_call
|
|
async def context_based_model(request: ModelRequest, handler) -> ModelResponse:
|
|
# 从 runtime context 读取配置
|
|
model_spec = request.runtime.context.model
|
|
model = load_chat_model(model_spec)
|
|
|
|
request = request.override(model=model)
|
|
return await handler(request)
|
|
|
|
|
|
class ReActAgent(BaseAgent):
|
|
name = "智能体 Demo"
|
|
description = "A react agent that can answer questions and help with tasks."
|
|
|
|
def __init__(self, **kwargs):
|
|
super().__init__(**kwargs)
|
|
self.graph = None
|
|
self.workdir = Path(sys_config.save_dir) / "agents" / self.module_name
|
|
self.workdir.mkdir(parents=True, exist_ok=True)
|
|
|
|
def get_tools(self):
|
|
return get_buildin_tools()
|
|
|
|
async def get_graph(self, **kwargs):
|
|
if self.graph:
|
|
return self.graph
|
|
|
|
# 创建 ReActAgent
|
|
graph = create_agent(
|
|
model=model,
|
|
tools=self.get_tools(),
|
|
middleware=[context_aware_prompt, context_based_model],
|
|
checkpointer=await self._get_checkpointer(),
|
|
)
|
|
|
|
self.graph = graph
|
|
logger.info("ReActAgent 使用内存 checkpointer 构建成功")
|
|
return graph
|