Append Tools dynamically to AgentExecutor after AgentExecutor object had been created

[tomask736]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

On server level:

#Define tools
summary = SummaryTool(
    ...
)

www_search = BingSearchAgent(
    ...
)

sql_search = CosmosDBQueryTool(llm=llm)

# Create the Agent
agent = create_openai_tools_agent(llm, [], CUSTOM_CHATBOT_PROMPT)
agent_executor = AgentExecutor(agent=agent, tools=[], callbacks=[logging_callback])


On specific HTTP request level:
```python
@app.post("/agent/invoke", response_model=Output, tags=["agent"])
async def agent_invoke(item: AgentInput, user_roles: str = Header(...)) -> Any:

    header_list = [item.strip() for item in user_roles.split(',')]

    if "group1" in header_list:
        index_name = "group1-index"
    elif "group2" in header_list:
        index_name = ["group1-index", "group2-index"]
    else:
        raise HTTPException(status_code=400, detail="Invalid value in User-Roles header.")

    doc_indexes=index_name if isinstance(index_name, list) else [index_name]

    doc_search = DocSearchAgent(
        ...
        indexes=doc_indexes,
        ...
    )

    tools = [doc_search, www_search, summary, sql_search]

    # Assign the tools to the Agent and AgentExecutor
    agent_executor.tools = tools

    brain_agent_executor = RunnableWithMessageHistory(
        agent_executor,
        get_session_history,
        ...
    )

### Description

I am using FastAPI to serve content and I would like to have one instance of Agent and AgentExecutor objects for the whole server. The reason is if I do these instances for each HTTP request from scratch, it takes a lot of time.

I would like to select available Tools based on header user_roles of actual REST API request, so we could provide different tools for different group of people.

I have checked documentation, discussions and also LangChain chat, but had not found anything. The provided idea of code is "invented" by LangChain chat, but is not working on line
` agent_executor.tools = tools`

giving error:
openai.BadRequestError: Error code: 400 - {'error': {'message': "Invalid 'tools': empty array. Expected an array with minimum length 1, but got an empty array instead.", 'type': 'invalid_request_error', 'param': 'tools', 'code': 'empty_array'}}


**Is is possible to append/assign Tools [] dynamically to Agent / AgentExecutor after respective objects had been created ?**
As of now, it seems it's possible only in constructor when creating given objects which slows down HTTP responses significantly as objects have to be created with each HTTP request separately from scratch.


### System Info

System Information
------------------
> OS:  Windows
> OS Version:  10.0.22631
> Python Version:  3.11.9 (tags/v3.11.9:de54cf5, Apr  2 2024, 10:12:12) [MSC v.1938 64 bit (AMD64)]

Package Information
-------------------
> langchain_core: 0.2.41
> langchain: 0.2.16
> langchain_community: 0.2.17
> langsmith: 0.1.135
> langchain_cli: 0.0.31
> langchain_cohere: 0.2.4
> langchain_experimental: 0.0.65
> langchain_mistralai: 0.1.13
> langchain_openai: 0.1.25
> langchain_text_splitters: 0.2.4
> langgraph: 0.2.37
> langserve: 0.2.3

Other Dependencies
------------------
> aiohttp: 3.10.5
> async-timeout: Installed. No version info available.
> cohere: 5.11.0
> dataclasses-json: 0.6.7
> fastapi: 0.115.0
> gitpython: 3.1.43
> gritql: 0.1.5
> httpx: 0.27.2
> httpx-sse: 0.4.0
> jsonpatch: 1.33
> langgraph-checkpoint: 2.0.1
> langgraph-sdk: 0.1.33
> langserve[all]: Installed. No version info available.
> numpy: 1.26.4
> openai: 1.51.2
> orjson: 3.10.7
> packaging: 24.1
> pandas: 2.2.2
> pydantic: 1.10.13
> pyproject-toml: 0.0.10
> PyYAML: 6.0.2
> requests: 2.32.3
> requests-toolbelt: 1.0.0
> SQLAlchemy: 2.0.35
> sse-starlette: 1.8.2
> tabulate: 0.9.0
> tenacity: 8.5.0
> tiktoken: 0.7.0
> tokenizers: 0.20.1
> tomlkit: 0.12.5
> typer[all]: Installed. No version info available.
> typing-extensions: 4.12.2
> uvicorn: 0.23.2