Overview
Build a bounded agent queue with explicit states:
- Intake. Create an issue from a user report, telemetry event, feature request, or human
observation.
- Triage. Decide how important the item is and whether it should advance. Keep prioritization under
human or clearly defined organizational control.
- Explore. Let an AFK agent inspect the codebase, investigate the problem, and describe possible
changes.
- Classify. Require structured output indicating whether the item is understood, whether it can be
implemented automatically, and where human input is required.
- Implement. Trigger a sandboxed agent only after the task has adequate scope and acceptance
criteria.
- Review. Run automated checks and produce a review recommendation.
- Decide. Auto-merge only classes of change that have earned that level of trust; otherwise present
the completed package to a human.
- Close and learn. Remove the item from the queue and capture any system improvement revealed by
the work.
This workflow can give the human a much richer starting point. Instead of receiving only a raw bug report, the human can receive the report, the codebase exploration, the proposed fix, the completed implementation, and a review summary.