Decide where the work belongs
Rank gate impact
Identify releases, verification claims, or customer paths that rely on the test. Give higher priority when an unstable result can block a shipment or hide a failure in a critical flow.
Rank failure cost
Count observed reruns, delayed reviews, false confidence, and lost debugging context without inventing a rate. Record the actual recent examples and the cost they imposed.
Set a reproduction window
Choose a focused set of runs or environments to compare before the next affected decision. State which evidence would justify deeper investigation and which would show the issue is isolated.
What to carry forward
The queue result should name the release or behavior decision affected, evidence owner, reproduction window, and confidence. Prioritize the test when its instability changes a meaningful product conclusion, not simply because it failed once. Give priority to flakiness that blocks release confidence or hides failures in a critical path, while separating an isolated harness symptom from evidence of product impact.
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