Error analysis and complex problem-solving specialist powered by Codex CLI. Use proactively when encountering errors, test failures, build failures, or unexpected behavior. Also use for complex debugging that requires deep reasoning. Automatically suggested by hooks when errors are detected.
Rare escalation authority for design arbitration, unblocking stuck problems, and final review of large changes — and for landing the resolution when handing it back down would just repeat a failure. Invoke sparingly — routine reviews belong to team-execute Phase 2 or /codex:adversarial-review, and routine…
Opus subagent for research, large-scale analysis, difficult implementation, and Codex delegation. Use when a task needs broad context, deep judgment, cross-cutting changes, or escalation from Sonnet.
Sonnet subagent for routine, well-scoped implementation. Use for approved plans, localized features and fixes, tests, refactoring, and mechanical file work with clear acceptance criteria.
Comprehensive onboarding for new or returning contributors. Scans repository artifacts (git history, .agents/STATE.md, project rules, skill catalog, DESIGN.md, research & library notes, checkpoints, agent-team logs) and synthesizes a GUIDE.md at the repository root summarizing what has been worked on, why, and how to…
Record a project design decision into .agents/docs/DESIGN.md through the shared typed writer. Use when the user says "record this", "add to design", "document this", "記録して", or asks what has been decided so far — and when a design decision has just been made and should not be lost.
Unified feature planning & implementation skill — replaces the old /add-feature and /start-feature skills (both trigger phrases still apply here). MODE=existing (formerly /add-feature): add a feature to an established codebase with Codex-first collaboration — Codex is consulted in every phase for scope analysis…
Time-boxed technical investigation/feasibility study with Codex-first multi-agent collaboration (Codex + Opus 4.6 + Agent Teams). Codex CLI is consulted in EVERY phase for question framing, feasibility analysis, and final evaluation. Phase 1: Frame the investigation question & constraints (Claude user interaction +…
Two-phase Agent Teams execution — replaces the old /team-implement and /team-review skills. Phase 1 IMPLEMENT (formerly /team-implement): parallel implementation with teammates per module/layer, file-ownership separation, and a shared task list with dependencies. Run after /feature plan approval. Phase 2 REVIEW…
Diagnose and plan fixes for errors/bugs with Codex-first multi-agent collaboration (Codex + Opus 4.6 + Agent Teams). Codex CLI is consulted in EVERY phase for deep code reasoning, hypothesis evaluation, and fix validation. Phase 1: Error reproduction & context gathering (Opus subagent 1M context + Codex initial…
Runs before the agent uses a tool for Edit and Write tool calls, executing check-codex-before-write.py via python3. From DeL-TaiseiOzaki/claude-code-orchestra.