# navmds MCP server

Research 7,400+ US doctors: search, semantic search, profiles, reviews & procedure pricing.

## Links
- Registry page: https://www.getdrio.com/mcp/io-github-mattlsmith-navmds
- Repository: https://github.com/mattlsmith/navmds
- Website: https://www.navmds.com

## Install
- Endpoint: https://www.navmds.com/api/mcp
- Auth: Not captured

## Setup notes
- Remote endpoint: https://www.navmds.com/api/mcp

## Tools
- search_doctors (Search doctors by procedure & location) - Find doctors by procedure/specialty, optionally filtered by US state and city. Uses NavMDs' pre-computed search index (fast, ranked by rating + review volume). Use this when you have a concrete procedure (e.g. 'rhinoplasty', 'botox', 'tummy tuck'). For open-ended natural-language queries, use semantic_search_doctors instead. Endpoint: https://www.navmds.com/api/mcp
- find_local_specialists (Curated 'best {procedure} doctors in {city}' guide) - Return NavMDs' curated city guide for a procedure: the ranked specialists (the city's OWN doctors first, then the nearest within an adaptive radius), the typical local price range, the average rating, the share offering free consultations, and the canonical navmds.com page URL to cite. This is the best tool for 'who are the best <procedure> doctors in <city>' questions — it mirrors NavMDs' editorial city pages. If no curated guide exists for that city+procedure, it says so; fall back to search_doctors with a nearby larger city. Endpoint: https://www.navmds.com/api/mcp
- semantic_search_doctors (Natural-language doctor search) - Search NavMDs' 7,400+ doctor directory with a natural-language query, e.g. 'board-certified facelift surgeon in Los Angeles with great reviews and free consults'. Powered by Gemini embeddings + cosine similarity over full doctor profiles. Best tool for open-ended or multi-attribute questions. Endpoint: https://www.navmds.com/api/mcp
- get_doctor (Get full doctor profile) - Fetch a doctor's complete NavMDs profile by slug (the id returned by the search tools, or the last path segment of a navmds.com/doctor/<slug> URL). Returns specialty, locations, ratings, overview, procedures, pricing, consultation fee, FAQ and verification status. Endpoint: https://www.navmds.com/api/mcp
- get_doctor_reviews (Get patient reviews for a doctor) - Return published (moderation-approved) patient reviews for a doctor by slug. Endpoint: https://www.navmds.com/api/mcp
- get_procedure_costs (Get pricing for a procedure) - Aggregate what a procedure costs across NavMDs doctors — combining patient-reported costs and practice-website pricing. Optionally narrow to a state/city. Returns min/median/max/average plus sample practice quotes. Endpoint: https://www.navmds.com/api/mcp
- list_procedures (List available procedures) - List the procedures/specialties covered by NavMDs, with the number of doctors for each. Useful for grounding a search. Endpoint: https://www.navmds.com/api/mcp
- list_locations (List covered locations) - List states/cities where NavMDs has doctors, with doctor counts. Useful for grounding a location filter. Endpoint: https://www.navmds.com/api/mcp
- search_community (Search NavMDs community discussions) - Search patient discussion threads on community.navmds.com — the NavMDs forum where people discuss recovery timelines, what procedures actually cost them, choosing a surgeon, and what to expect. Use this for lived-experience questions ('what is rhinoplasty recovery like', 'how much did people actually pay for a tummy tuck', 'how do I pick a surgeon') where the doctor-directory tools can't help. Returns thread titles, categories, excerpts and canonical URLs; pass an id to get_community_thread for the full discussion. Community posts are patient discussion, not medical advice. Endpoint: https://www.navmds.com/api/mcp
- get_community_thread (Get a full community discussion) - Fetch a complete NavMDs community thread by id (the id returned by search_community, or the trailing number of a community.navmds.com/t/<slug>/<id> URL) — the original question plus every reply, in order. Use after search_community when you need the actual answers rather than an excerpt. Community posts are patient discussion, not medical advice. Endpoint: https://www.navmds.com/api/mcp
- search (Search (connector-compatible)) - Search NavMDs doctors by free-text query. Returns a list of {id, title, url} results. Pass an id to the `fetch` tool to retrieve the full profile. (OpenAI connector-compatible.) Endpoint: https://www.navmds.com/api/mcp
- fetch (Fetch document (connector-compatible)) - Retrieve a full doctor profile by id (the slug returned from `search`). Returns {id, title, text, url, metadata}. (OpenAI connector-compatible.) Endpoint: https://www.navmds.com/api/mcp

## Resources
Not captured

## Prompts
- research_doctor - Research a doctor Pull a doctor's profile, reviews and pricing and summarize for a prospective patient. Arguments: name_or_slug
- patient_experience - What a procedure is actually like Answer a lived-experience question from real NavMDs community discussion, grounded with directory pricing. Arguments: question
- compare_doctors_for_procedure - Compare doctors for a procedure Find and compare the best doctors for a procedure in a location. Arguments: procedure, location

## Metadata
- Owner: io.github.mattlsmith
- Version: 1.0.0
- Runtime: Streamable Http
- Transports: HTTP
- License: Not captured
- Language: Not captured
- Stars: Not captured
- Updated: May 26, 2026
- Source: https://registry.modelcontextprotocol.io