Langchain MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 88 installs a week from npm. Last commit 19 Jul 2026.
LangChain/LangGraph/DeepAgents knowledge + LangGraph agent debugging (Polly-like)
Today is the operative word: we check Langchain MCP every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add langchain-mcp -- npx -y langchain-mcp
{
"mcpServers": {
"langchain-mcp": {
"args": [
"-y",
"langchain-mcp"
],
"command": "npx"
}
}
}
[mcp_servers.langchain-mcp]
command = "npx"
args = ["-y", "langchain-mcp"]
{
"mcpServers": {
"langchain-mcp": {
"args": [
"-y",
"langchain-mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"langchain-mcp": {
"args": [
"-y",
"langchain-mcp"
],
"command": "npx"
}
}
}
Inject deep domain knowledge into any AI agent
Persistent code knowledge graph + 9 composite MCP tools for AI coding agents.
LangChain Login and Token Agent
Intelligent code knowledge graph for AI coding agents — 71% cheaper, 72% faster
Governance layer for AI coding agents: knowledge-graph grounding, session audit, policy controls.
Repository knowledge graph MCP server for codebase understanding and debugging.
OAuth LangChain OpenAI Agent
Code knowledge graph over MCP: token-budgeted context for AI coding agents.
Answers built from our own checks of this server.