zhaoxi
b15eeb9e74
前端/DB 仍用 toolset 做逻辑分组管理，但传给 pydantic-ai Agent 时 把 toolset 内的 callable 展开为 tools=[] 扁平列表，MCP server 等 需要 toolset 语义的单独走 toolsets=[] 参数。解决工具"存在但调不了"的问题。 Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
114 lines
4.5 KiB
Python
114 lines
4.5 KiB
Python
# Copyright 2026 zhaoxi826
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
from pydantic_ai import Agent, RunContext
|
|
from pydantic import Field
|
|
from kilostar.adapter.model_adapter.agent_factory import AgentFactory
|
|
from kilostar.core.global_state_machine.model_provider.base_provider import Provider
|
|
from kilostar.utils.agent_model import ResponseModel, RequestModel, DepsModel
|
|
from kilostar.utils.ray_hook import ray_actor_hook
|
|
|
|
from kilostar.utils.logger import get_logger
|
|
|
|
logger = get_logger("worker_individual")
|
|
|
|
|
|
class WorkerIndividualResponse(ResponseModel):
|
|
"""Worker Individual 的输出模型，承载一次任务执行后的结果文本。"""
|
|
|
|
output: str = Field(..., description="Worker执行任务的输出结果")
|
|
|
|
|
|
class WorkerIndividualDeps(DepsModel):
|
|
"""Worker Individual 的运行期依赖，注入到 pydantic-ai Agent 的 RunContext。"""
|
|
|
|
task_event: dict
|
|
|
|
|
|
class WorkerIndividualInput(RequestModel):
|
|
"""Worker Individual 的输入模型，承载一次任务事件的入参。"""
|
|
|
|
task_event: dict
|
|
|
|
|
|
class BaseIndividual:
|
|
"""
|
|
Worker Individual 的基类
|
|
"""
|
|
|
|
def __init__(self, agent_config: dict):
|
|
self.agent_config = agent_config
|
|
self.agent_id = agent_config.get("agent_id")
|
|
self.agent: Agent | None = None
|
|
|
|
async def _init_agent(self, agent_name: str, system_prompt: str, toolsets=None):
|
|
"""根据 agent_config 拉起一个 pydantic-ai Agent 实例。
|
|
|
|
从 GlobalStateMachine 取出 Provider，按 agent_config 中的 provider_title
|
|
和 model_id 选择模型，加载工具集，并把 system_prompt 注册为动态提示词。
|
|
若调用方未显式提供 ``toolsets``，会自动从全局状态机拉取配置的工具集。
|
|
|
|
Args:
|
|
agent_name: Agent 的人类可读名称，用于日志与展示。
|
|
system_prompt: 该 Agent 的基础系统提示词，会和 task_event 拼接成动态提示词。
|
|
toolsets: 显式传入的外部工具集；为 ``None`` 时会自动按配置拉取。
|
|
"""
|
|
from kilostar.utils.mcp_helper import get_all_tools_and_toolsets_for_scope
|
|
from kilostar.core.global_state_machine.gsm_snapshot import fetch_snapshot
|
|
|
|
global_state_machine = ray_actor_hook(
|
|
"global_state_machine"
|
|
).global_state_machine
|
|
provider_title = self.agent_config.get(
|
|
"provider_title", "openai"
|
|
) # default fallback
|
|
model_id = self.agent_config.get("model_id", "gpt-4o") # default fallback
|
|
toolset_ids = self.agent_config.get("tools", None)
|
|
|
|
# 直读快照，避开 actor RPC 单线程串行
|
|
snapshot = await fetch_snapshot(gsm_actor=global_state_machine)
|
|
provider: Provider = snapshot.providers.get(provider_title)
|
|
if provider is None:
|
|
raise ValueError(f"Provider {provider_title!r} 未注册")
|
|
agent_factory = AgentFactory()
|
|
|
|
if toolsets is None:
|
|
tools, toolsets = await get_all_tools_and_toolsets_for_scope(
|
|
agent_name, toolset_ids=toolset_ids
|
|
)
|
|
else:
|
|
tools = []
|
|
|
|
self.agent = agent_factory.create_agent(
|
|
provider=provider,
|
|
model_id=model_id,
|
|
output_type=WorkerIndividualResponse,
|
|
system_prompt=system_prompt,
|
|
deps_type=WorkerIndividualDeps,
|
|
agent_name=agent_name,
|
|
tools=tools,
|
|
toolsets=toolsets,
|
|
)
|
|
|
|
@self.agent.system_prompt
|
|
async def dynamic_prompt(ctx: RunContext[WorkerIndividualDeps]):
|
|
"""把基础 system_prompt 与本次 task_event 拼接成最终动态提示词。"""
|
|
prompt = system_prompt + "\n\n"
|
|
prompt += f"=== 当前任务上下文 ===\n{ctx.deps.task_event}\n"
|
|
return prompt
|
|
|
|
async def run(self, task_event: dict) -> dict:
|
|
"""执行一次任务，需要由子类按自身策略实现。"""
|
|
raise NotImplementedError("子类必须实现 run 方法")
|