Skip to content

Talala Content API

A read-only JSON view of the reference library Talala publishes on the web: blog posts, recipes, calculators, quizzes, guides, and biomarkers. Every item carries the canonical page URL it came from.

Base URL

https://www.talala.app — every response is JSON, served over HTTPS, and readable from any origin.

Endpoints

GET /api

Lists every endpoint in this API along with the version, the OpenAPI description, and the documentation page.

GET /api/health

Liveness signal for the API. Returns 200 with a status of "ok" while the API is serving requests.

GET /api/blog

Every published Talala blog post, newest first, as metadata plus the canonical article URL.

GET /api/recipes

Every published recipe with its tags, timing, and serving facts, plus the canonical recipe URL.

GET /api/resources

Calculators, quizzes, guides, and biomarker references from the resources library, each tagged with its resource type and canonical URL.

Machine-readable description

The OpenAPI 3.1 description lives at https://www.talala.app/openapi.json. Automated clients can find it without reading this page: Talala publishes an API catalog at https://www.talala.app/.well-known/api-catalog, as described in RFC 9727, with the API anchored at https://www.talala.app/api.

No key, no sign-up

Every endpoint is a public GET. There is no authentication, and no rate limit beyond ordinary edge protection. Please be reasonable — responses are static, so caching them on your side costs you nothing.

Responses are cached, not live

Calculators, guides, biomarkers and recipes are compiled into the site, so those collections only change when Talala redeploys. Blog articles are published separately and refresh within about five minutes. Responses are cached at the edge for a day and revalidated in the background. Health is never cached.

Metadata, not full text

Collections return summaries plus the canonical URL of the page each item came from. The full article, recipe method, and calculator itself live on those pages. This matches the Content-Signal declaration in robots.txt: indexing and AI answers that link back are welcome, model training is not.

Versioning

The current version is 1.0.0. Fields may be added without notice; anything removed or renamed comes with a major version bump and a new path.