legacy-plannerlisted

# /legacy-planner — Former planning chain This is the preserved legacy workflow. The active autonomous pipeline is `goals`; do not auto-trigger this skill for ordinary planning requests. Orchestrate the planning skills **in order**, carrying context forward through each, with an explicit invariants gate before the spec. This is the repeatable "plan the next effort and hand it off" loop. Do **not** write any application code in this skill — it produces planning artifacts only. ## Your place in the fleet loop The fleet is a pipeline on the project's board, one skill per stage, each stage run by a **different model** so that no stage ever reviews its own work: ``` Planned → Agent Ready → Coding → Debugger Ready → Debugging → Review Ready → Reviewing → Done └ /legacy-planner ─┘ └──── /legacy-coder ────┘ └──── /legacy-debugger ────┘ └──── /legacy-reviewer ────┘ ``` ### Stage-parent session The default planner is Opus 5 in the Claude Code harness using the Claude subscription. GPT-5.6 Luna may dispatch it as a headless child after all planning decisions are supplied, or the human may run it in a visible Claude Code session. Because the grill is interactive, a headless child cannot ask the user. `/legacy-planner` may use one-level helpers only when it is the independent top-level stage parent; a native child launched from another session must not spawn nested children. **Your board moves:** create GitHub issues in the project's **`Planned`** Status, then move each pr