Technical checker · API description quality
Measure whether AI agents can understand your API
A parseable specification is the beginning, not the finish. Santos examines whether an agent can discover the document, select an operation, construct valid input, anticipate output and errors, and understand access or payment requirements.
What the assessment covers
Valid
Parseable OpenAPI 3.x or Swagger 2.0 with a canonical server.
Typed
Inputs, outputs, examples, required fields, and stable operation identifiers.
Operational
Auth, payment, errors, limits, settlement, and retry behavior.
Canonical discovery
The document should be advertised from public HTML, Link metadata, or machine guidance and identify the canonical server. Conflicting or stale specifications reduce confidence even when each file parses.
Operations an agent can select
Stable operationId values, meaningful summaries, tags, request schemas, required fields, response schemas, and examples reduce guesswork during tool construction.
Access and failure behavior
Authentication, x402 payment, rate limits, idempotency, timeouts, non-2xx responses, and retry guidance should be explicit so an agent can act without unsafe trial and error.
Frequently asked questions
Does every website need MCP or OpenAPI?
No. They matter when a site exposes callable services. Santos marks non-applicable checks instead of penalizing an informational website.
What is an Agent Readiness audit?
It checks the public technical signals that help agents find information, interpret structured meaning, and invoke applicable machine-readable capabilities.