菜单
Pick the right worker class for each Issue. Use every time you assign an Issue; wrong routing wastes tokens and creates rework.
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。
Break a Company Goal into 3-10 atomic Issues. Use when a Goal arrives with no pre-existing plan and you need to produce a shippable work list.
Ship the smallest change that solves the Issue. Use on every implementation task; scope creep is the default failure mode of executors.
Run build/test/lint and read the actual output before claiming done. Use on every implementation task, at the end.
Start implementation with a red test. Use on any bugfix or new behavior — not on refactors or style changes.
Report research results in a form other agents can act on. Use at the end of every Explorer session.
Find files and code patterns fast. Use for every codebase question that starts with "where is X" or "how is Y used".
| name | dispatch-by-role-strengths |
| description | Pick the right worker class for each Issue. Use every time you assign an Issue; wrong routing wastes tokens and creates rework. |
Workers are not interchangeable. Misrouting an Issue costs tokens and often produces the wrong output shape.
Before creating a Reviewer sub-issue, check the executor's adapter capability. Claude-Code-family adapters self-review; for those, skip the Reviewer and post the audit comment Skipped PetAgent Reviewer: executor adapter self-reviews (spec §3.4). PetAgent-native executors always need the separate Reviewer pass.
Issues that do not touch the same files can run in parallel on different executors. Declare dependencies explicitly in the Dispatch Summary so the board can schedule correctly.
