Skill Claude Code
Multi-step git commit workflow. Gathers state, triages sensitive content and secrets, groups changes into logical units, runs the project's test/lint gate, proposes a plan for approval, and commits with conventional messages and no AI attribution. Use when changes need to be committed — when the user says "commit"…
Skill Claude Code
Closes out a just-finished implementation slice and emits a self-contained handoff prompt for a fresh session — gates that the slice landed (else stops), re-anchors the plan's stale refs against current code, and harvests what building revealed. Use when the user says "next-slice", "hand off the next slice", "close…
Skill Claude Code
Harden an existing implementation plan via a fixed N-round cross-model critique→fold loop. A codex (GPT) subprocess critiques the plan against the real code; a claude (opus) subprocess rules on each finding and folds accepted fixes into a copy. Use when the user explicitly asks to harden, stress-test, clash, or…
Skill Claude Code
Interviews the user relentlessly about a plan, design, or proposal until shared understanding is reached, walking the decision tree branch by branch and resolving dependencies one at a time. Use when the user asks to stress-test, pressure-test, grill, interrogate, or challenge a plan, design, or proposal; or when the…
Skill Claude Code
Synthesize a decisions record and the current codebase into a pure implementation plan written to disk. Use when the user asks to turn a plan, spec, or set of decisions into an implementation plan, or says "to-plan this", "write a plan", "make a plan from this".
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: