Cursor rule Cursor
Never attribute commits to AI tools — human author only.
Cursor rule Cursor
Never attribute commits to AI tools — human author only.
Skill Codex
Scaffold the agent-kit — a shared Cursor .mdc rules layout with symlinks across repos.
Skill Codex
Run a benchmark to measure skill quality, response time, and token efficiency for aicrew skills.
Skill Codex
Wrap up a session in Codex with summary, tests, risks, and a commit message proposal.
Skill Codex
Scaffold a local Cursor extension for a multi-tool terminal panel (Claude Code + Codex + Gemini).
Skill Codex
Run the aicrew /dev pipeline in Codex (intake -> research -> design -> TDD -> tests -> security -> audit -> infra -> conclude).
Skill Codex
Run the aicrew /fix fast bug fix flow in Codex (triage -> root cause -> TDD -> verify).
Skill Codex
Generate a compact cross-tool handoff block from current session state for switching tools or models.
Skill Codex
Audit the aicrew harness health and completeness in Codex.
Skill Claude CodeCodex
Run or guide aicrew install — first-time global setup of skills, hooks, and MCP wiring.
Skill Codex
Disable lean mode and restore verbose, full-prose output for this session.
Skill Codex
Run the aicrew /quick scout-first, Karpathy-guarded lightweight flow in Codex.
Skill Codex
Set session label for state checkpoints — makes .ai/state/ files multi-tool safe and easy to resume.
Skill Claude CodeCodex
Show what aicrew has installed — /Agents, hooks, commands, Codex skills, and project .ai/skills/.
Skill Codex
Re-enable terse/caveman output if it was previously disabled with aicrew-normal.
Skill Codex
Update aicrew global and project skills in Codex (runs aicrew update and optional project generation).
Skill Codex
Re-run aicrew install to pick up new skills from the package (merge only — preserves user edits).
Skill Codex
Generate 3 materially different implementation options with trade-offs before coding, using project-specific brainstorm overrides when available.
Skill Codex
Default low-token operating mode for aicrew — terse output and context-economy read policy. Use /lean on to boost or re-enable after /normal.
Command Claude Code
Benchmark aicrew skill quality, trigger accuracy, token efficiency, and pipeline timing.
Command Claude Code
Generate 3 materially different implementation options with trade-offs before coding.
Command Claude Code
Use at session end to save project learnings and produce a commit message.
Command Claude Code
Use when starting any development task — bug fix, feature, or refactor.
Command Claude Code
Use for a fast bug fix with minimal ceremony (3 questions, TDD, done).
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: