菜单
Use when authoring a new skill, editing an existing skill, upgrading a flexible skill to rigid, or the user says "/write-skill", "create a skill", "new skill for X", or "make this a skill". Fires before any new `SKILL.md` is committed.
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
Use when the user says "/deep-review", "deep review", "thorough review", or wants the deepest possible code review before pushing a branch.
Use when the user says "/harness-update", "update the harness", "upgrade agent-harness", "pull the latest harness", or after seeing an UPGRADE_AVAILABLE notice. Pulls latest harness files into this project, preserves project-specific configuration and local-only skills, and walks the user through any conflicts.
Use when the user wants to audit or understand the agent harness itself — its hooks, skills, agents, or commands — or asks "what's in this harness", "how does this harness work", or "harness overview".
Use when the user has a one-off task that slots into the current sprint and needs a quick written plan — lighter than /plan-sprint. Reads existing plans for context, uses plan mode, writes a focused plan file.
Use when the user says "/build", hands off a sprint-plan document path, or asks to execute a written plan end-to-end. Branches, implements, tests, verifies in browser, commits incrementally, and prepares the PR. Auto-fires /lg-* skills when plan steps involve LangGraph/LangChain agent work.
Use when the user says "review migrations", "db review", "check schema changes", "safe to migrate?", or before deploying schema changes to production.
| name | write-skill |
| description | Use when authoring a new skill, editing an existing skill, upgrading a flexible skill to rigid, or the user says "/write-skill", "create a skill", "new skill for X", or "make this a skill". Fires before any new `SKILL.md` is committed. |
| user-invocable | true |
| tier | rigid |
| kind | process |
Override: see
CLAUDE.md§ Instruction precedence. The user is principal; this skill is advisory.
Writing a skill IS test-driven development applied to process documentation. If you didn't watch a subagent fail without the skill, you don't know what the skill is for. Merges superpowers' writing-skills with .claude/docs/harness-principles.md and skills/CONVENTIONS.md. Violating the letter is violating the spirit.
NO RIGID SKILL SHIPS WITHOUT (1) A NAMED FAILURE MODE FROM A SUBAGENT
BASELINE, (2) A RATIONALIZATION TABLE QUOTING VERBATIM EXCUSES, (3) A
TRIGGERS-ONLY DESCRIPTION, (4) A DECLARED TERMINAL STATE, AND (5) AN
`eval.yaml` DECLARING AT LEAST ONE TRAJECTORY EVAL (OR ONE INVOCATION
EVAL FOR SUBAGENT-ONLY SKILLS).
Flexible/util skills are exempt from (1), (2), (5) — never from (3). Wrote the body before baselining? Delete the rationalization table; don't keep imagined rows "as reference." (§3, §11.) Missing eval.yaml? bin/skill-eval --validate-strict fails.
CLAUDE.md or a hook (checklist.md)./skill-baseline against a subagent without the skill. Capture verbatim excuses._template-rigid/{TEMPLATE.md,eval.yaml}. Counter the baseline rationalizations verbatim → rationalizations.md. Translate GREEN trajectory → eval.yaml. Re-run with the skill; subagent should cite the section that prevented the failure.bin/test-frontmatter and bin/skill-eval --validate both pass. Then run /skill-eval <new-skill> to execute the first trajectory eval against a fresh subagent — confirms the eval contract is satisfiable, not just well-formed.VERSION, commit, push, open a draft PR.References: skill-md-template.md, checklist.md, .claude/docs/skill-eval-spec.md.
eval.yaml trajectories from imagination — no docs/skill-baselines/ source link.eval.yaml (bin/skill-eval --validate-strict fails).@-loading a sibling (force-loads context) — use **REQUIRED SUB-FILE:** Read foo.md.<update-check> block, missing override pointer, frontmatter omits tier or kind.All of these mean: stop. Move the work back into RED. Expanded list: red-flags.md.
REQUIRED SUB-FILE: Read rationalizations.md if you find yourself making excuses. The verbatim-excuse-to-reality table is seeded from superpowers' writing-skills baseline and flagged for re-baseline in this harness.
Before committing:
name. bin/test-frontmatter passes.Use when, lists concrete triggers, contains zero workflow summary.<update-check> block present immediately after frontmatter.rationalizations.md, Self-Review Checklist, Terminal State all present. Every rationalization row traces to a file under docs/skill-baselines/.eval.yaml present with at least one trajectory eval (or one invocation eval for subagent-only skills). bin/skill-eval --validate passes.wc -w SKILL.md < 500 (target) / < 700 (hard ceiling). Overflow → siblings.VERSION bumped.Cannot check all boxes? You skipped this skill. Restart from the top.
Hooks (different contract), commands (thin wrappers), memory entries (see CLAUDE.md § Learnings), and doc-only edits (no baseline; still bump VERSION patch if user-visible).
Terminal state is a draft PR to agent-harness containing the new/edited skill, baseline transcripts under docs/skill-baselines/, the populated eval.yaml, and a VERSION bump. bin/test-frontmatter and bin/skill-eval --validate must pass. Do NOT invoke /ship — the draft is on purpose. Do NOT use the new skill on real work until the user reviews the rationalization table and the PR merges.