Command
Audit the project's memory tree for drift — orphans, duplicates, broken relations, stale references, schema gaps. Read-only by default; proposes fixes via propose-confirm-commit. Sibling of /review for memory rather than code.
Command
Scaffold a new Claude Code subagent under agents/ — interview-driven, forces job-to-be-done, pre-flight questions, anti-patterns, and debugging workflow. Writes one .md file with full structure and a settings/MEMORY.md side-effect-free preview.
Command
Generate a vendor-neutral AGENTS.md by projecting this Claude-Code instruction corpus — CLAUDE.md tree, architectural rules, warning/feedback landmines, canonical commands — into a flat file other coding agents (Codex, Cursor, Aider, …) can read, then interviewing only for the gaps the corpus can't know.
Command
Scaffold a new Claude Code hook end-to-end — recipe-driven hook file, payload fixtures, runner, and settings.json registration with diff preview.
Command
Scaffold a new MCP server project under mcps/ — interview-driven, supports TypeScript and Python, simple tool servers and API wrappers. Writes project files, installs dependencies, builds, and registers in /.claude.json with diff preview.
Command
Goal-directed orchestration & convergence for concurrent work — decompose one goal into interdependent units, place each (shared tree / worktree / serialize), fan out via dispatch, keep on track at dispatch, then verify-and-converge into one coherent result.
Command
Draft a PR title + body to a clean contract (Summary + Test plan + Closes-line over any existing template) and hand you a ready-to-run gh pr create command — never opens the PR itself.
Command
Review a GitHub Pull Request — fetch diff via gh, analyze for correctness/bugs, design/patterns, PR hygiene, and security. Present structured findings locally. No GitHub posting.
Command
Triage a PR's unresolved review comments into a checklist, then walk each to one of three outcomes — Act (route the code change to /dispatch or /orchestrate), Skip/defer, or Note (capture a decision for the reply you'll write). Never drafts or posts replies.
Command
Run the pre-push pre-flight checklist before pushing — commits-to-ship summary, branch-name / commit-hygiene / AI-attribution / debug-artifact / hook-bypass checks; stops on any flag.
Command
Prime the session with architectural rules (universal / language / domain / project) relevant to the current task before writing code.
Command
Write an end-of-session recap — request / investigated / learned / completed / next-steps — with propose-confirm-commit flow and optional promotion of Learned items.