🦜 🔗 LangChain MCP Client

What is 🦜 🔗 LangChain MCP Client?

A simple model context protocol client that demonstrates the use of MCP server tools by a LangChain ReAct agent. It seamlessly connects to any MCP servers, supports any LangChain-compatible LLM for flexible model selection, and offers a CLI for dynamic interactions.

How to use 🦜 🔗 LangChain MCP Client?

Install with pip install langchain_mcp_client (Python 3.11+). Create a .env file with API keys for your LLM and configure llm_mcp_config.json5 with your LLM parameters, MCP servers, and optional example queries. Then launch the CLI (e.g., make cli or the appropriate command for your operating system).

Key features of 🦜 🔗 LangChain MCP Client

• Converts MCP tools into LangChain-compatible List[BaseTool].
• Supports parallel initialization of multiple MCP servers.
• Works with any LangChain-compatible LLM.
• CLI-based dynamic conversation interface.
• Example queries can be provided in the config file.

Use cases of 🦜 🔗 LangChain MCP Client

• Letting a LangChain agent use tools from external MCP servers.
• Querying a Jupyter MCP server to create notebooks and visualizations.
• Combining multiple MCP services through natural language prompts.

FAQ from 🦜 🔗 LangChain MCP Client

What Python version is required?

Python 3.11 or higher.

How do I configure the client?

Create a .env file with your LLM API keys and a llm_mcp_config.json5 file that specifies the LLM, MCP servers, and optional example queries.

Can I use any LLM with this client?

Yes, any LangChain-compatible LLM can be used.

How do I run the client after configuration?

Launch the CLI — the README shows examples using make cli (commands may vary by OS).