MCP server that parses LangGraph Python codebases into a structural graph model and layers local semantic search on top.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"langgraph-context-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
<div align="center"> # LangGraph Context MCP **Ask your LangGraph codebase what it actually does — structurally, not by grepping.** [](LICENSE) [](https://www.python.org/downloads/) [](#status--limits) [](https://github.com/KarimHabib100/LangGraph-Context-MCP/actions/workflows/ci.yml) [](https://modelcontextprotocol.io) </div> --- LangGraph Context MCP is a Model Context Protocol server that parses a LangGraph Python codebase into a structural graph model — nodes, edges, conditional routing, tool bindings — and layers local semantic search on top, so an AI coding assistant can answer questions about your agent's architecture without reading every file. Structural answers come from parsing the actual graph definition, so they are exact rather than inferred. Everything runs locally: no API key, and no network access after the embedding model downloads once. ``` "how does execution get from clarify_with_user to the final report?" clarify_with_user → write_research_brief → research_supervisor → final_report_generation (branch at clarify_with_user, routed by the node itself) ``` ## Requirements - **Python 3.11 or newer** - **[uv](https://docs.astral.sh/uv/getting-started/installation/)**, only if you use the `uv` install path below. `pip` works without it. - Roughly **500 MB of disk** for the embedding model, downloaded once on first use. - No API key. No database server. No network access at query time. ## Install > **Not yet on PyPI.** The package name is registered as available but publication happens at > launch, so the two commands below are the correct forms and will work once it is published. > Until the…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.