Codex hit its usage limit. Resuming safely is harder than clicking Continue.

Description and use

When Codex hits its usage limit, resuming long tasks is harder than clicking Continue. Continua preserves interrupted tasks, verifies the environment before resuming, and avoids duplicate submissions, though unattended recovery after a real usage-limit cycle remains unproven.

Problem

When Codex usage limits are hit, interrupted long coding tasks must be resumed safely, but detecting the interruption, verifying the environment, and avoiding duplicate submissions is difficult.

Outcome

The host-retry fix in Continua is implemented; the engine suite passed 336 tests and 27 native-controller checks. The patched build safely cancels stale continuations when newer user activity appears. However, unattended recovery through a fresh real usage-limit cycle remains unproven.

Workflow

• Detect Codex task interruption and save task metadata and scope
• Wait for usage limit reset or monitor usage availability
• Before resuming, verify that the live Codex session matches the saved task and environment
• If no submission has been made, retry environment verification within a bounded recovery window
• If a submission may already exist, reconcile its outcome before any further submission
• If newer user activity is detected, cancel the stale continuation
• Resume the task by sending a continuation when all checks pass

Evidence

The host-retry fix is implemented and installed. The engine suite passed 336 tests, and 27 isolated native-controller checks passed.

Source

A timer answers when to check again. It does not establish that a particular task is safe to resume.

Source

The current implementation checks the original live writer and the rollout file it actually has open. It also checks the task metadata and project scope.

Source