Skip to content
OpenAPI for AI agents

Check if your OpenAPI spec is ready for agent use.

An OpenAPI file can be valid and still be risky for AI agents. TimeProofs AgentReady checks whether endpoints, schemas, descriptions and dangerous actions are explicit enough before agent-facing use.

Scan OpenAPIPreview report

What makes OpenAPI agent-ready?

  • Operation names are specific enough for tool selection.
  • Descriptions explain when to use and when not to use endpoints.
  • Dangerous actions require confirmation.
  • Parameters have clear bounds, enums and examples.
  • Error responses help agents stop, correct or retry safely.

Common failure patterns

  • POST /refunds without confirmation rules.
  • DELETE endpoints without recovery notes.
  • Broad exports without filters or scope limits.
  • Free-text fields where enums should exist.
  • Generic 500 errors with no corrective guidance.

AgentReady output

The scanner returns an AgentReady Score, severity counts, endpoint findings, recommended fixes, a Markdown report and a reusable agentready.json contract.

The analysis runs locally in the browser. It does not call your uploaded API endpoints.

Limit

TimeProofs AgentReady does not guarantee that an AI agent will never fail. It identifies structural risks that may cause AI agents to misuse APIs, tools or MCP servers.

AgentReady at a glance

OpenAPI readiness for AI agents means checking whether operations, schemas, descriptions and dangerous actions are explicit enough before agent-facing use. AgentReady is a product and candidate open standard for static pre-deployment analysis. It is not a runtime firewall, independent certification, official standards-body standard or guaranteed-safety system.

Community
AgentReady Community is free and available for local browser, CLI and GitHub Action use.
Pro
AgentReady Pro is planned and not purchasable.
Language
English is the current canonical language. No translated alternate route is published for this page.
Boundary
The page explains static analysis and pre-deployment review, not runtime protection or certification.
Sources

AGENTREADY_RULE_CODES.mdparse-openapi.jsgenerate-agentready-json.jsSEO_GEO_AI_FIRST_REQUIREMENTS.mdGLOBAL_STANDARD_SITE_PROGRAM.md