メニュー
Create, edit, or audit the Codex-focused Superpowers bundle. Use Codex's built-in skill-creator for general skill authoring.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Execute an existing plan when implementation is authorized and dependent outcomes need tracking; skip short routine checklists.
Mandatory newcomer-safe router for every conversation. Classify by action, impact, and named risk; add workflows only when their descriptions match.
Create a durable plan when explicitly requested, required by the High-risk tier, or needed for resumable dependency coordination; skip routine native planning.
Resolve ambiguous product or system work before implementation. Use when materially different interpretations would change scope, architecture, or consequential outcomes; skip precise tasks.
Complete an authorized commit, push, integration, PR, branch, or worktree action while preserving user state. Use after relevant verification when the user or repository requires delivery.
Evaluate incoming code-review feedback against repository evidence. Keep evaluation-only requests read-only; implement supported fixes only when authorized.
| name | writing-skills |
| description | Create, edit, or audit the Codex-focused Superpowers bundle. Use Codex's built-in skill-creator for general skill authoring. |
Read the nearest AGENTS.md, inspect the working tree, and audit each directly affected skill. For bundle-wide policy, compare every direct skill's description and body; load support resources only when activation or an intended edit depends on them.
Front-load one discriminative trigger per invocation branch in each description. Keep only non-obvious procedure in the body; make branch-only resources conditional and directly linked.
Remove duplication, stale prose, model-relative no-ops, and lifecycle rules owned elsewhere. Preserve authorization, user-work, security, and evidence guardrails.
Consult Codex's built-in skill-creator when creating structure, changing metadata format, or resolving a validation question; routine wording revisions do not need its full guide reloaded.
Validate every changed skill:
python3 "${CODEX_HOME:-$HOME/.codex}/skills/.system/skill-creator/scripts/quick_validate.py" "$SKILL_DIR"
Run bash skills/superpowers/check-context-budget.sh, relevant script checks, and git diff --check.
Forward-test only when inspection and validators cannot establish activation or high-risk behavior. Use the smallest discriminating scenario and do not leak the expected answer. Report measured size, behavior changes, checks, and untested risk.