Agent-readiness
Agent-readiness is the degree to which AI agents can find and understand an API from its public documentation surface — the pages, specs, and entry-point files an agent actually fetches. An agent-ready API presents a machine-legible discovery path and documentation a model can operate from, and that is what determines whether an agent chooses it when building an integration.
AI agents are becoming a primary consumer of APIs. When a developer asks an agent to build something, the agent reads public documentation, decides which API it can operate, and quietly picks one — no sales call, no evaluation spreadsheet. There are two ways to lose that moment, and they happen in order: the agent never finds a usable version of the docs at all, or it finds them and cannot operate from them — misreading auth, guessing at endpoints, inventing capabilities the API does not have.
Agent-readiness is a property of the public surface, measured before execution. It is distinct from runtime API quality: authentication, successful calls, reliability, and SDK behavior are separate questions. The term gives teams a definition of done for the request that otherwise arrives with none — "make us AI-ready" — by naming the thing that can actually be observed and improved: what an agent can find and understand from what you publish.
Discry measures agent-readiness as the Discry Score, always decomposed into discovery and comprehension. Discovery asks whether a machine-legible surface exists at all — llms.txt, AGENTS.md, an OpenAPI spec, MCP presence — on a plain HTTP fetch, the way agents actually arrive. Comprehension is measured behaviorally: real models are quizzed on the live docs and graded mechanically against ground truth cited to the documentation itself. Everything is measured before a single API call, and every profile publishes its receipts.
What does an AI agent make of your API?
Find out in about a minute — no signup.
Discry your API — free