跳到正文
cccskills
免费在 GitHub 公开

hook-updater

Research-backed hook refresh workflow for updating existing hooks with TDD checkpoints and settings.json registration validation.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md3.3 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Hook Updater

Overview

Use this skill to refresh an existing hook safely: research current best practices, compare against current implementation, generate a TDD patch backlog, apply updates, and verify ecosystem integration including settings.json registration.

When to Use

  • Reflection flags stale or malfunctioning hook behavior
  • EVOLVE determines hook capability exists but quality is outdated
  • User asks to audit/refresh an existing hook
  • Hook exit code or stdin protocol issues detected

The Iron Law

Never update a hook blindly. Every refresh must be evidence-backed, TDD-gated, and integration-validated.

Workflow

Step 0: Evaluate Current State

  1. Read the hook file and understand its purpose
  2. Check settings.json registration (which events, which matchers)
  3. Run existing tests if they exist
  4. Identify issues: wrong API pattern, missing registration, broken requires

Step 1: Research Best Practices

  1. Read .claude/hooks/CLAUDE.md for hook conventions
  2. Review similar hooks in the same category for patterns
  3. Check .claude/rules/hooks.md for hook rules
  4. Use Skill({ skill: 'research-synthesis' }) if external research needed

Step 2: Generate Patch Backlog

  1. List specific changes needed (TDD-style: test first, then implement)
  2. Prioritize by risk: registration fixes > exit code fixes > logic improvements
  3. Estimate scope: <30 lines = auto-apply, >30 lines = plan mode

Step 3: Apply Updates

  1. Write or update tests first (RED phase)
  2. Apply hook changes (GREEN phase)
  3. Refactor for clarity (REFACTOR phase)
  4. Use Edit tool — never rewrite the entire file

Step 4: Verify Integration

  1. Confirm hook is registered in .claude/settings.json for correct events
  2. Verify hook exit codes: 0=allow, 2=block, 1=error (SE-03)
  3. Verify stdin JSON parsing follows the standard protocol
  4. Run pnpm lint:fix && pnpm format
  5. Run hook-specific tests

Step 5: Record

  1. Log changes via MemoryRecord if significant
  2. Update CHANGELOG.md entry

Domain-Specific Validation

  • Hook MUST be registered in settings.json under appropriate event
  • Hook MUST use stdin JSON protocol (not programmatic function exports)
  • Hook MUST exit 0 (allow) or 2 (block) — never exit 1 for blocking
  • Hook MUST wrap body in try/catch, exit 0 on unexpected errors (SE-03)
  • Hook require() paths MUST be file-relative (../../lib/...), not CWD-relative

Anti-Patterns

  • Updating a hook without checking its settings.json registration
  • Changing exit codes without understanding the 0/1/2 semantics
  • Converting a stdin-protocol hook to programmatic API (or vice versa)
  • Removing error handling to "simplify" the hook

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Ensure accessibility in UI components including semantic HTML, ARIA attributes, keyboard navigation, and WCAG 2.2 AA compliance.

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

Use when you want to improve response quality through meta-cognitive reasoning. Applies 15+ reasoning methods to reconsider and refine initial outputs.

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

N-round opposing-stance debates for trade-off analysis. Assigns pro/con roles to agents, runs structured debate rounds with quality scoring, and produces a moderator synthesis with confidence-rated recommendation. Generalizable to architecture, technology, security, and design decisions.

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

Force adversarial code review stance that eliminates confirmation bias — reviewer must find issues or re-analyze

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

Creates specialized AI agents on-demand when no existing agent matches a request. Use when the Router cannot find a suitable agent for a task. Enables self-evolution by generating persistent agents.

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

LLM-as-judge evaluation framework with 5-dimension rubric (accuracy, groundedness, coherence, completeness, helpfulness) for scoring AI-generated content quality with weighted composite scores and evidence citations

中文简介正在准备中。当前显示原文说明。

oimiragieo/agent-studio422026年7月14日 更新

查看 oimiragieo 的全部技能

报告此技能的问题