Skip to content

Beket API

Measure how AI surfaces answer questions about your brand, and read the results programmatically. Create a target, configure it, run a measurement, read what came back.

Quickstart

The whole loop end to end in curl, TypeScript, or Python — from creating a target to reading a report. Start here.

Core concepts

Audit, prompts, run, snapshot, report — what each one is and how they nest.

Reading the report

Headline metrics, winnable prompts, the content roadmap — one section at a time, through GraphQL.

Authentication

Where keys come from, what they reach, and why a forbidden resource returns 404.

Errors

The envelope, every status code, and which failures are worth retrying.

This documentation is written to be read by software as well as by people.

  • /docs/openapi.json — the OpenAPI 3.1 document. Feed it to a client generator, an editor, or an agent. Also served live and public at GET /api/v1/openapi.json — no credentials required, so a client can read the contract before it holds a key.
  • /docs/llms.txt — every page, its title and its description, in one fetch.
  • Any page, as markdown — append .md to a URL for the source with none of the site chrome. /docs/guides/quickstart.md is this site’s quickstart, unstyled.

The reference is generated from the API’s own request and response schemas, and a test fails the build if the published pages drift from them. What you read here is what the server validates.