MCP Server
Connect ChatGPT or another compatible Model Context Protocol client to Quantified Self, approve only the read-only data categories you want it to use, and ask questions about your training history without exposing your entire account by default.
Training Questions
The server exposes bounded tools for discovery and analysis. Every result is projected through the permissions you approve instead of returning raw account documents.
Discover available metrics, inspect day, week, or month trends, and query ready Training analysis such as load, readiness, intensity, durability, and sport-specific evidence.
Find activities by time or nearby start and end positions, then inspect bounded summaries, laps, swim lengths, MTB jumps, app links, and selected numeric Sports Lib metrics.
Review normalized sleep sessions plus daily, weekly, or monthly aggregates without exposing provider payloads, raw stage intervals, or raw physiological samples.
Search saved routes near coordinates or a place, and inspect bounded route metrics, simplified geometry, segment start and end positions, and waypoint coordinates.
Access Boundaries
MCP access uses an authorization flow designed for external clients. The client receives only the scopes you approve and can be disconnected from Connections.
Activity metrics, individual activity details, sleep summaries, and saved routes are independent grants. Adding a new category requires another explicit approval.
MCP clients cannot create, edit, or delete activities, routes, sleep records, dashboard settings, or Training state. Revocation blocks future access.
Raw files, raw streams, provider credentials, and device provenance are excluded. Activity details and route scopes can include exact coordinates, so authorize location access only for clients you trust.
Direct-coordinate searches stay within Quantified Self. For a place-name search, only the supplied location text is sent to Mapbox for geocoding—not your activity, route, account, or unrelated prompt data.
FAQ
It gives a compatible external client a bounded, read-only way to discover and query approved Quantified Self activity metrics, Training analysis, individual activity details, sleep summaries, and saved routes.
Yes. Add https://quantified-self.io/mcp as a custom app endpoint in ChatGPT, sign in to Quantified Self, and approve the read-only permissions you want to grant. ChatGPT is an external client with its own privacy and retention practices.
No. The MCP server is read-only. It cannot write dashboard settings, activities, routes, Training state, or sleep records.
No. Original files, raw streams, provider payloads, credentials, and storage paths are excluded. Results use bounded summaries and safe projections for the specific tool and permission.
Only if you approve individual activity details or saved-route access. Those scopes can return exact start, end, jump, route, or waypoint coordinates and may reveal sensitive places such as a home, workplace, or frequent trailhead.
Next Step
Create an account, connect the training sources you use, and authorize only the MCP permissions needed for the questions you want to ask. Review or revoke completed client connections at any time.