What it is
A public archive that snapshots popular free APIs every day and records exactly what changed in their response schemas — and when.
APIs change silently. A field gets renamed, a type flips from number to string, a nested object disappears — and your integration breaks at 2am. CtrlZAPI is a public, version-controlled archive of what popular free APIs actually return, snapshotted every day, with a field-level changelog of every structural change.
How the pipeline works
- Fetch. Every day at 06:00 UTC, a GitHub Actions workflow runs
node scripts/snapshot.js. It fetches one endpoint per tracked API — sequentially, with a 1-second delay between requests and a 15-second timeout each. - Extract.
scripts/schema.jsconverts each JSON response into a canonical schema tree: objects become field maps, arrays merge the schemas of their elements, primitives becomestring/number/boolean/null. Key order never affects the result. - Diff. The new schema is hashed and compared with yesterday's. On a mismatch,
diffSchemasproduces field-level changes —added,removed,type_changed— addressed by dot-paths likelicense.spdx_idortags[]. - Publish. Snapshots, changelogs, and the index are committed back to
data/as plain JSON. This site reads those files directly — no backend, no database, no cost.
If a fetch fails, the previous snapshot is kept untouched and the API is marked error until the next successful run.
APIs currently tracked
CtrlZAPI tracks 33 production-grade, keyless APIs across 12 categories:
- Weather — Open-Meteo, Open-Meteo Air Quality, Open-Meteo Geocoding, USGS Earthquakes, Sunrise-Sunset
- Finance — Kraken, Coinbase, Fear & Greed Index, Frankfurter FX, ExchangeRate-API
- Geo — REST Countries, Zippopotam, Nager Public Holidays, IP-API
- Books — Open Library, Gutendex
- Art — Met Museum, Art Institute of Chicago
- Dev — GitHub REST API, npm Registry, PyPI, Hacker News, HN Algolia API, Stack Exchange
- Space — Open Notify (ISS), Open Notify (Astronauts)
- Reference — Hipolabs Universities
- Media — TVMaze, Jikan (MyAnimeList)
- Food — TheMealDB, TheCocktailDB
- Sports — Jolpica F1
- Jobs — Remotive
All are keyless and free. Endpoints are fixed (not parameterized per run) so day-to-day diffs reflect real API changes, not different inputs.
Add your own API
- Add an entry to
config/apis.json:{ "id": "my-api", "name": "My API", "category": "Reference", "description": "What this API does.", "baseUrl": "https://api.example.com", "headers": { "Accept": "application/json" }, "endpoints": [ { "path": "/v1/things", "label": "List things" } ] } - Run
node scripts/snapshot.jslocally to create the seed snapshot. - Commit. The next scheduled run picks it up automatically.
Rules: the API must need no key, the endpoint must be stable (same input every day), and responses should be JSON under ~1MB. category groups the API on the dashboard; headers is optional and only needed when an endpoint requires a special request header (e.g. Accept: application/json).
Local development
No build step, no dependencies. To refresh the data: node scripts/snapshot.js (Node 18+). To preview the site: serve the repo root with any static server, e.g. npx serve . or python3 -m http.server, then open index.html.