# Clair

> Clair is a semantic news API that matches meaning, not keywords: describe what you're watching for and get the news that means it. Scored, explainable, open-core.

## Docs

- [Introduction](https://clair.im/docs/api-reference/introduction.md): Base URL, authentication, the embedding-space contract, completeness guarantees, and error codes for the v1 data API.
- [Match news](https://clair.im/docs/api-reference/match.md): Score the corpus against a text or vector query, or omit the query to page the corpus chronologically.
- [List sources](https://clair.im/docs/api-reference/sources.md): The news outlets Clair actively serves, by name: the valid values for the sources and exclude_sources filters.
- [How it works](https://clair.im/docs/how-it-works.md): The stateless-vector model, the shared embedding space, and the two-stage match behind every score.
- [Clair News API](https://clair.im/docs/index.md): A news-intelligence data API. Define the entities you care about locally; get every article Clair linked to them, deterministically.
- [Quickstart](https://clair.im/docs/quickstart.md): From zero to matched news in five minutes, no install, no vectors required.

## OpenAPI Specs

- [openapi](https://clair.im/docs/api-reference/openapi.json)