工具使用

虽然大语言模型（LLM）可以完成各种各样的任务，但在需要全面专业知识的领域中，它们可能表现不佳。此外，LLM 还可能遇到幻觉问题，而这些问题很难靠自身解决。

因此，我们需要使用一些工具来帮助 LLM 完成任务。

备注

在 DB-GPT 智能体中，大多数 LLM 都支持工具调用，只要其自身能力不是太弱即可。 （例如 glm-4-9b-chat、Yi-1.5-34B-Chat、Qwen2-72B-Instruct 等）

编写工具

有时候，LLM 可能无法直接完成计算任务，因此我们可以编写一个简单的计算器工具来帮助它们。

from dbgpt.agent.resource import tool

@tool
def simple_calculator(first_number: int, second_number: int, operator: str) -> float:
    """Simple calculator tool. Just support +, -, *, /."""
    if isinstance(first_number, str):
        first_number = int(first_number)
    if isinstance(second_number, str):
        second_number = int(second_number)
    if operator == "+":
        return first_number + second_number
    elif operator == "-":
        return first_number - second_number
    elif operator == "*":
        return first_number * second_number
    elif operator == "/":
        return first_number / second_number
    else:
        raise ValueError(f"Invalid operator: {operator}")

为了测试多个工具，我们再编写一个工具来帮助 LLM 统计目录中的文件数量。

import os
from typing_extensions import Annotated, Doc

@tool
def count_directory_files(path: Annotated[str, Doc("The directory path")]) -> int:
    """Count the number of files in a directory."""
    if not os.path.isdir(path):
        raise ValueError(f"Invalid directory path: {path}")
    return len(os.listdir(path))

将工具封装为 ToolPack

大多数情况下，你可能有多个工具，因此可以将它们封装为一个 ToolPack。 ToolPack 是工具的集合，你可以用它来管理你的工具，智能体可以根据任务需求从 ToolPack 中选择合适的工具。

from dbgpt.agent.resource import ToolPack

tools = ToolPack([simple_calculator, count_directory_files])

在智能体中使用工具

import asyncio
import os
from dbgpt.agent import AgentContext, AgentMemory, LLMConfig, UserProxyAgent
from dbgpt.agent.expand.tool_assistant_agent import ToolAssistantAgent
from dbgpt.model.proxy import OpenAILLMClient

async def main():

    llm_client = OpenAILLMClient(
        model_alias="gpt-3.5-turbo",  # 或其他模型，例如 "gpt-4o"
        api_base=os.getenv("OPENAI_API_BASE"),
        api_key=os.getenv("OPENAI_API_KEY"),
    )
    context: AgentContext = AgentContext(
        conv_id="test123", language="en", temperature=0.5, max_new_tokens=2048
    )
    agent_memory = AgentMemory()
    agent_memory.gpts_memory.init(conv_id="test123")

    user_proxy = await UserProxyAgent().bind(agent_memory).bind(context).build()

    tool_man = (
        await ToolAssistantAgent()
        .bind(context)
        .bind(LLMConfig(llm_client=llm_client))
        .bind(agent_memory)
        .bind(tools)
        .build()
    )

    await user_proxy.initiate_chat(
        recipient=tool_man,
        reviewer=user_proxy,
        message="Calculate the product of 10 and 99",
    )

    await user_proxy.initiate_chat(
        recipient=tool_man,
        reviewer=user_proxy,
        message="Count the number of files in /tmp",
    )

    # dbgpt-vis 消息信息
    print(await agent_memory.gpts_memory.app_link_chat_message("test123"))
    
if __name__ == "__main__":
    asyncio.run(main())

输出结果如下：

--------------------------------------------------------------------------------
User (to LuBan)-[]:

"Calculate the product of 10 and 99"

--------------------------------------------------------------------------------
un_stream ai response: {
  "thought": "To calculate the product of 10 and 99, we need to use a tool that can perform multiplication operation.",
  "tool_name": "simple_calculator",
  "args": {
    "first_number": 10,
    "second_number": 99,
    "operator": "*"
  }
}

--------------------------------------------------------------------------------
LuBan (to User)-[gpt-3.5-turbo]:

"{\n  \"thought\": \"To calculate the product of 10 and 99, we need to use a tool that can perform multiplication operation.\",\n  \"tool_name\": \"simple_calculator\",\n  \"args\": {\n    \"first_number\": 10,\n    \"second_number\": 99,\n    \"operator\": \"*\"\n  }\n}"
>>>>>>>>LuBan Review info: 
Pass(None)
>>>>>>>>LuBan Action report: 
execution succeeded,
990

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
User (to LuBan)-[]:

"Count the number of files in /tmp"

--------------------------------------------------------------------------------
un_stream ai response: {
  "thought": "To count the number of files in /tmp directory, we should use a tool that can perform this operation.",
  "tool_name": "count_directory_files",
  "args": {
    "path": "/tmp"
  }
}

--------------------------------------------------------------------------------
LuBan (to User)-[gpt-3.5-turbo]:

"{\n  \"thought\": \"To count the number of files in /tmp directory, we should use a tool that can perform this operation.\",\n  \"tool_name\": \"count_directory_files\",\n  \"args\": {\n    \"path\": \"/tmp\"\n  }\n}"
>>>>>>>>LuBan Review info: 
Pass(None)
>>>>>>>>LuBan Action report: 
execution succeeded,
19

--------------------------------------------------------------------------------

在上面的代码中，我们使用 ToolAssistantAgent 来选择并调用合适的工具。

更多细节？

在上面的代码中，我们使用 tool 装饰器来定义工具函数。它会将函数封装为一个 FunctionTool 对象。而 FunctionTool 是 BaseTool 的子类，BaseTool 是所有工具的基类。

实际上，工具是 DB-GPT 智能体中一种特殊的资源。你可以在资源章节中了解更多细节。