Implementing the tool_select_tool tool

[arkonchen]

Checked other resources

• This is a feature request, not a bug report or usage question.
• I added a clear and descriptive title that summarizes the feature request.
• I used the GitHub search to find a similar feature request and didn't find it.
• I checked the LangChain documentation and API reference to see if this feature already exists.
• This is not related to the langchain-community package.

Package (Required)

• langchain
• langchain-openai
• langchain-anthropic
• langchain-classic
• langchain-core
• langchain-cli
• langchain-model-profiles
• langchain-tests
• langchain-text-splitters
• langchain-chroma
• langchain-deepseek
• langchain-exa
• langchain-fireworks
• langchain-groq
• langchain-huggingface
• langchain-mistralai
• langchain-nomic
• langchain-ollama
• langchain-perplexity
• langchain-prompty
• langchain-qdrant
• langchain-xai
• Other / not sure / general

Feature Description

Whether or not the tool selection tool will be realized.I think this is a very good tool, unlike LLMToolSelectorMiddleware.
And, could the type of request.tools be changed to dict to make it easier to find tools? Instead of looping through a list to find tools

Use Case

https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool#when-to-use-tool-search

Proposed Solution

No response

Alternatives Considered

No response

Additional Context

No response

[towseef41]

@arkonchen

There’s already a tool picker (LLMToolSelectorMiddleware) that asks a model to choose the best tools before the main call, so the “select relevant tools” job is covered; if you just want quicker lookup by name we can add a small helper like request.tools_by_name while keeping the current list to avoid breaking anything, and a separate “tool search” tool only makes sense if you need something callable mid-run or a different selection strategy I am happy to add the helper or sketch that separate tool if you want.

[austinmw]

@towseef41 Could we get an embedding based version instead of an LLM-based version too?