legacy-debuggerlisted

# /legacy-debugger — Former debugging chain This is the preserved legacy workflow. The active autonomous pipeline is `goals`; do not auto-trigger this skill for ordinary debugging requests. The **skeptic** of the fleet. Where `/legacy-planner` plans and `/legacy-coder` builds, `/legacy-debugger` deliberately tries to break what was built — on a different model from the one that wrote it, which is the whole point: the misreading that produced a bug also produces a confident review of that bug, and only an outside context breaks that correlation. It makes sure the repo has a **personalized auditor agent** (creating it if absent, reusing it if present), then runs that agent as a **systematic debugger**: read the context, hunt bugs through four nets, fix them test-first, move the ticket on. It **composes** `/legacy-planner` and `/legacy-coder` rather than reinventing them — bugs are framed with `/legacy-planner`'s ticket format (invariant + `Verification-command`) and fixed with `/legacy-coder`'s test-first loop. The ticket trail is the audit log. ## Your place in the fleet loop ``` Planned → Agent Ready → Coding → Debugger Ready → Debugging → Review Ready → Reviewing → Done └ /legacy-planner ─┘ └──── /legacy-coder ────┘ └──── /legacy-debugger ────┘ └──── /legacy-reviewer ────┘ ``` ### Stage-parent session The default debugger is GPT-5.6 Luna in a separate top-level Codex session using **maximum reasoning effort**. It may use auditor helpers inside that independent