You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[AW Top 10] 06 Stop codex resume retries from invalid item IDs #66490
Codex runs retried as resume fail with an invalid_request_body error saying input item IDs must begin with fc, and every retry hits the same error. The engine driver then exits without a terminal safe output.
Implementation scope
Reproduce the resume failure in the codex harness, then either stop resuming when the error is a deterministic request-body rejection or sanitize the replayed item IDs. Report the cause in the failure issue.
Done when
A harness test shows an invalid_request_body resume error is not retried identically.
The failure issue names the request-body rejection rather than a generic driver failure.
Why now
Two issues show the identical error and two more show the same workflow failing with driver exit one. The latter pair is plausibly the same cause but is not verified.
No corroborating AW discussion; evidence comes from the source issues.
Unchanged AW sources close only after this summary is completed. Newer source activity and not-planned retirement do not trigger source closure. Assigned summaries are frozen; unassign to allow reclustering.
Activity