Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Use before implementing anything - dispatches fresh subagent per task with parallel dual review and a spec gate. As an orchestrator, you must never implement anything yourself.
Facilitate real subagent roundtable meetings for uncertain futures, difficult decisions, product or architecture brainstorming, pre-mortems, and explicit disagreement during spec writing. Use during `writing-specs` before approach selection when multiple plausible paths need pressure-testing, stakeholder perspectives conflict, or potential product issues need to be surfaced. Use only when subagent use has already been approved for the current workflow; otherwise stop. Do not use for routine implementation tasks, settled decisions, or simulated roleplay.
Parallel dual review via subagents — spec compliance gates code quality. Guides evaluation of reviewer feedback.
You MUST use this before any implementation work. Clarifies requirements, explores approaches, validates the design, and writes the approved spec before planning.
Use when you have a spec or requirements for a multi-step task, before touching code
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
Short reminder of the double-SDD Codex workflow and instruction priority