菜单
Run project onboarding — reads START_HERE.md and executes the protocol there. Auto-detects new project vs. existing codebase, interviews you (new) or reads the code and asks only what it can't answer (existing), fills docs, builds the backlog. Usage: /start
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Save-and-commit checkpoint — verifies docs are updated, runs available linting/tests, then creates a WIP commit so no work is lost before pausing. Usage: /checkpoint [optional: short description of current state]
Orchestrate a multi-agent task — analyzes dependencies, builds a wave execution plan, coordinates with the project manager, creates a feature branch, and runs specialist agents in parallel and sequential waves. Usage: /orchestrate <task description>
Multi-agent code review — architectural drift check, test coverage audit, and implementation quality review scoped to recently changed files. Usage: /review [optional: branch or file path scope]
Project health overview — renders a structured status card from TODO.md, recent git history, open PRs, and PRD open questions. Read-only. Usage: /status
Pre-release quality pass — checks backlog readiness, runs QA, updates docs, and verifies the CI/CD pipeline before shipping. Usage: /release [optional: version label e.g. v1.2.0]
Pull the latest agent definitions, skills, hooks, rules, and templates from the upstream orchestrated-project-template repository into this project's .claude/ directory, with a diff summary and confirmation before applying. Usage: /sync-template
| name | start |
| description | Run project onboarding — reads START_HERE.md and executes the protocol there. Auto-detects new project vs. existing codebase, interviews you (new) or reads the code and asks only what it can't answer (existing), fills docs, builds the backlog. Usage: /start |
Read START_HERE.md and execute the onboarding protocol defined there in full.
