Decide where the work belongs
Rank the action blocked by policy
Name whether the response blocks a read, save, upload, verification, or recovery action. Separate a background refresh from a person unable to complete a durable workflow. Give higher priority to limits that leave state uncertain or force unsafe manual retries.
Group reach by scope and traffic
Compare actor, credential, route, resource, window, concurrency, request mix, client, and release. Distinguish one noisy actor from a shared resource limit affecting unrelated work. Record confirmed reach and untested fairness separately, especially where a reset is not visible.
Choose a bounded limit slice
Select immediate investigation, a scope or retry-handling improvement, or collection of one missing reset comparison. State what changes priority, such as critical work denied while capacity is available or retries amplifying demand. Keep a complete policy redesign separate from the affected route.
What to carry forward
Produce a queue decision tied to blocked workflow, actor and resource reach, reset behavior, fairness, and evidence quality. Include whether retry is safe. The next engineer should have one rate-limited route and a clear reason its policy behavior deserves capacity.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow