Lookup MCP servers in the official MCP Registry by name or URL.
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":"mcp-lookup","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.
<p align="center"> <img src="https://raw.githubusercontent.com/websentry-ai/mcp-lookup/main/assets/logo.svg" alt="mcp-lookup logo" width="96" height="96"> </p> <h1 align="center">mcp-lookup</h1> <p align="center"> Lookup MCP servers across the <a href="https://registry.modelcontextprotocol.io">official MCP Registry</a>, <a href="https://smithery.ai">Smithery</a>, and <a href="https://pulsemcp.com">PulseMCP</a> from one CLI and Python library. Stdlib-only, no runtime dependencies. </p> ## Why Tools that need to enrich raw MCP server detections — e.g. a security dashboard that catalogs which MCP servers developers are running, an IDE assistant that needs canonical install metadata, or an SBOM pipeline — typically resort to HTML-scraping third-party catalogs because most registries don't expose a clean JSON API. `mcp-lookup` wraps the three main MCP catalogs (`registry.modelcontextprotocol.io`, `api.smithery.ai`, and `api.pulsemcp.com`) behind a tiny, stable interface. It gives you: - **Canonical server metadata** by name or repo URL: description, publisher namespace, version, packages (`npm`/`pypi`/`docker`), and remote transports (`stdio`/`sse`/`streamable-http`). - **Three-source aggregation**: `search` fans out to all three sources **in parallel** and returns merged results, each labeled by `_source`. `get` prefers the registry (canonical), then falls back to Smithery, then PulseMCP. - **Parallel search for low latency**: search latency is bounded by the slowest source (~1.2s) instead of the sum of all three (~5s). One slow or unavailable source never blocks the others. - **Authoritative tool lists**: Smithery entries include the server's `tools[]` with input schemas, which the registry does not expose. - **Community coverage**: PulseMCP adds GitHub star counts, package download counts, and long-tail community servers not (yet) in the official registry. - **URL → name resolution**: hand it a `https://github.com/owner/repo` (or a vendor domain) and every …
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.