generated: '2026-09-19'
method: searched
source: https://policycheck.tools/.well-known/agent.json
docs: https://policycheck.tools/docs#rate-limits
limit_count: 1
summary: >-
  Two provider statements disagree. The docs say the free endpoints are "currently free with no enforced rate
  limits"; the agent card's X-API-Key scheme description says "Free tier: 100 requests/minute. Contact
  api@policycheck.tools for higher limits." The 100/minute figure is the only numeric limit the provider
  publishes and is recorded as the limit; no rate-limit response headers are documented or were observed, and
  no exhaustion status code is published.
rate_limits:
- name: Free tier
  scope: per-key
  limit: 100
  window: 60s
  metric: request
  burst: null
  applies_to: [all endpoints (as stated in the agent card securitySchemes.apiKey description)]
  source: https://policycheck.tools/.well-known/agent.json
  quote: 'Free tier: 100 requests/minute. Contact api@policycheck.tools for higher limits.'
  condition: Stated for API-key holders; the docs say the anonymous /api/check and /api/a2a endpoints have no enforced limit during beta.
  domains: [policycheck.tools]
headers:
  documented: []
  observed_on_live_400: [cache-control, content-type, strict-transport-security, x-matched-path, x-vercel-cache, x-vercel-id, access-control-allow-origin]
  note: No RateLimit-*, X-RateLimit-* or Retry-After header on any observed response.
exhaustion:
  status: undocumented
  note: No 429 behaviour is documented.
caching_as_throttle:
  note: 'Results are cached by seller domain for 24 hours server-side; repeated requests for the same seller return the cached result. Providing policy_text bypasses the cache.'
paid_endpoint:
  note: POST /api/x402/analyze is metered by payment ($0.03 per request) rather than by a rate limit.