Is your API ready for AI agents?
Paste an OpenAPI spec URL (or the spec itself). Get a 0–100 AI-readiness score in ~30 seconds, with concrete fixes ranked by impact.
or try:
What we measure
D1
Description Coverage & Quality
What share of operations are described, and how usable are those descriptions for an LLM writing integration code?
D2
Example Coverage
Do request bodies, response schemas, and complex parameters carry concrete examples agents can mimic?
D3
Naming Consistency
Predictable case style, resource-noun paths, plural conventions, and uniform parameter names across endpoints.
D4
Auth Clarity
Are security schemes declared, described, and consistently referenced per operation — with scopes explained?
D5
Error Model
Are 4xx/5xx responses documented with a consistent, typed error schema agents can parse and react to?
More dimensions coming — see roadmap.