SuperAPI AI Directory

api-rate-limit-quota-review

Deployment & operations

View repository

Updated 10-16 02:20

About this skill

API Rate Limit Quota Review

Use this skill to convert a API quotas, rate limits, fair use, developer experience, billing alignment, and abuse controls question into a concrete artifact with owners, gates, metrics, and recovery paths.

Workflow

  1. Identify API consumer, authentication principal, endpoint cost, criticality, tenant blast radius, pricing tier, and abuse pattern.
  2. Read references/api-rate-limit-quota-patterns.md.
  3. Separate rate limit, quota, concurrency cap, payload cap, spend cap, fair-use review, and entitlement gate.
  4. Define limit units, headers, error shape, retry/backoff, dashboards, alerts, upgrade path, support override, and migration policy.
  5. Produce quota model, state machine, endpoint decision table, event schema, and developer communication checklist.

When not to use

  • Do not use for generic advice the base model already handles without this skill's specific artifact contract.

Guardrails

  • Do not use a single global limit when endpoint cost, tenant size, and abuse risk differ materially.
  • Do not return vague 429s without stable headers, retry guidance, documentation, and observability.
  • Do not let quota bypass billing, entitlement, security, or tenant isolation controls.
  • Do not silently change limits for existing developers without migration notice and support path.

Output format

API context:
Principal / endpoint cost / tenant blast radius:

Limit and quota plan:
| Area | Decision | Evidence | Risk | Owner |
| --- | --- | --- | --- | --- |

Developer contract:
- <item> -> <policy, metric, edge case, support note>

Override and migration policy:
- <trigger> -> <action, communication, owner>