FitTrack MCP
A read-only MCP server for fitness tracking, providing public app info and protected weight entries from Supabase.
README
FitTrack MCP
A read-only MCP server built with XMCP, Supabase Auth, and HTTP transport.
Tool
get-fittrack-info returns a fixed payload containing the app name, service
status, and sample tracking categories. It does not require authentication and
does not read or write external data.
get-recent-weight-entries is a protected, read-only tool that selects the
authenticated user's records from public.fittrack_weight using a month, an
exact date, an exact weight in kilograms, or a combination of date and weight.
When both inputs are omitted, it defaults to the current UTC month. It refuses
requests without a verified Supabase OAuth bearer token and relies on Supabase
RLS to enforce auth.uid() = user_id.
Results can be sorted by date or weight, ascending or descending. Without
sorting inputs, the query defaults to created_at descending.
get-recent-waist-entries applies the same protected, read-only filtering and
sorting behavior to public.fittrack_waist. It accepts a month or exact date,
an exact waist measurement in centimeters, and optional sorting by date or
waist. Without sorting inputs, it defaults to created_at descending.
get-recent-meal-entries applies protected, read-only filtering and sorting to
public.fittrack_meals. It accepts a month or exact date, a case-insensitive
food-description fragment, exact calories, and optional sorting by date,
calories, time, or food. Without sorting inputs, it defaults to
created_at descending.
The server publishes OAuth Protected Resource Metadata at
/.well-known/oauth-protected-resource. Supabase Auth is the OAuth 2.1
authorization server, while the MCP server remains the resource server.
Requirements
- Node.js 20 or newer
- npm
Local development
npm install
npm run dev
The MCP endpoint is available at http://localhost:3001/mcp.
Supabase connectivity safety test
Copy .env.example to .env.local and provide the project URL, an
sb_publishable_... key, and the canonical MCP endpoint URL. Never use a
secret or service-role key.
npm run test:db-read
The test performs an anonymous, read-only request against
public.fittrack_weight without returning row data. It passes only when RLS
hides every row from the anonymous role.
Authentication
Configure Supabase Auth with:
- OAuth 2.1 Server enabled
- Dynamic OAuth application registration enabled
- Site URL
https://fittrack.taimoorahmed.com - Authorization path
/oauth/consent
The web application owns the consent page. The MCP server verifies supplied
access tokens through Supabase Auth before passing them to the read-only
database client. Only OAuth-issued tokens containing a client_id are accepted.
Set the production resource URL to the exact endpoint used by MCP clients:
MCP_RESOURCE_URL=https://fittrackmcp.vercel.app/mcp
Build
npm run build
npm start
Deploy to Vercel
Import this Git repository into Vercel or run:
vercel deploy
After deployment, the public endpoint is https://<deployment>.vercel.app/mcp.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.