Summary
Settles F1 arguments with the public record instead of recall.
Seven tools carve the dataset along the lines people actually ask about — a driver, a constructor, a circuit, a season — which is why questions land without anyone naming a table. The data is F1DB under CC BY 4.0, imported and re-exposed here; the project is unofficial and unaffiliated with the Formula 1 companies.
What it is
A Python MCP server over F1 data, built from the open F1DB dataset imported into PostgreSQL and exposed as tools for a chat client.
What you get
- Seven entry points into the data: `circuit`, `constructor`, `driver`, `grand prix`, `manufacturer`, `race` and `season`
- Historical records rather than a live timing feed — the dataset is the public F1DB record
- Queries optimised against a relational import, instead of parsing files per question
Requirements
`pip install mcp-f1data`, then add a server entry that runs `python -m mcp-f1data`. The documented setup targets Claude Desktop.
Setup effort
One command — pip install mcp-f1data