Kumkuat API reference
This reference is generated from the API's own OpenAPI 3.1 specification, the same document that drives the Swagger UI at https://api.kumkuat.ai/docs, trimmed to the operations meant for integrations. The full document also carries the routes the Kumkuat app uses for its own sign-in flow, debug endpoints and pilot features that are switched off for most workspaces; those are left out here and still listed in the raw specification at https://api.kumkuat.ai/openapi/openapi.json.
Authentication
Send a workspace API key in the X-API-Key header on every request. Mint keys under Workspace → Integrations → API keys in the app; a key is bound to one workspace and acts with that workspace's rights.
curl -H "X-API-Key: $KUMKUAT_API_KEY" https://api.kumkuat.ai/api/v1/personas
Groups
| Group | Covers |
|---|---|
| Workspace | The current workspace and its extracted tag vocabularies |
| Personas & audiences | Personas, grounding, refinements and voice profiles, reactions and interactions, audience groups and summaries |
| Documents & content | Upload, ingest, query and tag documents; the content-writing endpoints; executive and theme insights |
| Signals & themes | Signal templates, signals and runs |
| Auto tags | Tag definitions, aliases and normalisation |
| Reports | Custom reports, sections, generation, PDF, email and scheduling |
Two areas sit behind a feature that Kumkuat enables per workspace, and their pages say so: Reports needs Custom reports, and the persona why this matters endpoint needs Why this matters. Without the feature the API answers 403.
Each endpoint page shows parameters, request body, response schemas and a request panel you can run against the API with your own key. See REST API for conventions, and the API how-to guides for worked examples: a first call, scoring a draft against audiences, integrating a content creation platform, building scorecards and chat, media pitching, message creation, and managing an audience library.