Plan AQE Work

1. Read AGENTS.md, the objective, relevant research, affected implementation, tests, configuration, and public boundaries.
2. Define observable outcomes, non-goals, constraints, assumptions, and decisions that require human approval.
3. Decompose work into independently verifiable deliverables. Map dependencies, consumers, migrations, compatibility boundaries, and rollback points.
4. Select the smallest useful specialist fleet with references/planning-framework.md. Give every selected specialty a concrete output; do not summon the whole fleet.
5. Arrange phases so contracts, safety controls, and durable tests precede or accompany implementation. Check CLI/MCP and Claude/Codex parity when shared capabilities cross those surfaces.
6. Identify work that can run in parallel and a clear critical path. Do not use unsupported precision for time estimates; size work by risk and dependency.
7. Attach acceptance criteria and the narrowest verification command to each deliverable. Include security, resilience, performance, accessibility, or data-integrity gates only when triggered by the actual scope.
8. End with prioritized milestones, risks and mitigations, open decisions, and a recommended first vertical slice.

Use Ruflo for persistent memory, learned routing, hooks, or multi-agent coordination only when its operational benefit exceeds its setup overhead. Treat Claude agent definitions as domain references; translate their workflows into available Codex actions rather than copying Claude-only tool syntax.