跳到正文
cccskills
免费在 GitHub 公开

dependency-auditor

Audit npm dependencies for security vulnerabilities, outdated packages, and unused dependencies. Use when checking for security issues, updating packages, or cleaning up dependencies.

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。

Dependency Auditor

Instructions

When auditing dependencies:

  1. Run security audit
  2. Check for outdated packages
  3. Find unused dependencies
  4. Analyze bundle size impact
  5. Review and update

Security Audit

# NPM audit
npm audit

# Get JSON output for processing
npm audit --json

# Fix automatically (safe fixes only)
npm audit fix

# Force fix (may have breaking changes)
npm audit fix --force

# PNPM
pnpm audit

# Yarn
yarn audit

Check Outdated Packages

# NPM
npm outdated

# Interactive update
npx npm-check-updates -i

# Update all to latest
npx npm-check-updates -u
npm install

# Check specific package
npm view <package> versions

Find Unused Dependencies

# Using depcheck
npx depcheck

# With details
npx depcheck --detailed

# Ignore patterns
npx depcheck --ignores="@types/*,eslint-*"

Common False Positives

Depcheck may flag these as unused when they're actually needed:

  • @types/* packages (used by TypeScript)
  • ESLint/Prettier plugins (referenced in config)
  • PostCSS plugins (referenced in config)
  • Next.js plugins
  • Babel presets

Analyze Bundle Size

# For Next.js
npx @next/bundle-analyzer

# General purpose
npx source-map-explorer dist/**/*.js

# Check package size before installing
npx package-phobia <package-name>

# Compare alternatives
npx bundlephobia-cli compare lodash ramda

Dependency Review Checklist

Security

  • No critical/high vulnerabilities
  • Dependencies actively maintained
  • No known malicious packages
  • Lock file committed

Freshness

  • No major version behind (unless intentional)
  • Security patches applied
  • Deprecated packages replaced

Cleanliness

  • No unused dependencies
  • No duplicate packages (check lock file)
  • devDependencies vs dependencies correct

Update Strategies

Conservative (Recommended)

# Update patch versions only
npm update

# Update specific package
npm install package@latest

Aggressive

# Update everything
npx npm-check-updates -u
npm install
npm test

Interactive

npx npm-check-updates -i

# Options:
# a - update all
# space - toggle selection
# enter - apply selected

Package.json Cleanup

{
  "dependencies": {
    // Runtime dependencies only
  },
  "devDependencies": {
    // Build/test tools only
  },
  "peerDependencies": {
    // For libraries only
  },
  "optionalDependencies": {
    // Platform-specific (rare)
  }
}

Lock File Best Practices

  1. Always commit lock files (package-lock.json, pnpm-lock.yaml, yarn.lock)
  2. Use npm ci in CI/CD (not npm install)
  3. Regenerate if corrupted: delete lock file + node_modules, reinstall
  4. Single lock file per project (don't mix package managers)

Automated Monitoring

# .github/dependabot.yml
version: 2
updates:
  - package-ecosystem: "npm"
    directory: "/"
    schedule:
      interval: "weekly"
    open-pull-requests-limit: 10
    groups:
      dev-dependencies:
        dependency-type: "development"

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

Audit websites for accessibility issues and WCAG compliance. Use when checking accessibility, fixing a11y issues, or ensuring WCAG compliance.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

Deploy a 2-layer parallel agent hierarchy for large, parallelizable work — big refactors, multi-file migrations, codebase-wide audits, bulk generation. A top-tier commander (Fable or Opus) orchestrates the swarms; the user picks a power level (Max Power / Heavy / Balanced / Economy) that sets the Opus/Sonnet/Haiku model mix per layer. Layer 1 is 3-50+ specialist agents, each with its own full context window; Layer 2 is 2+ sub-agents per member. Includes git safety, tiered sizing, a pre-deploy gate, phantom-completion checks, and multi-wave follow-up.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

Use to test whether an AI agent, prompt, or skill actually works before or after it ships — build a small eval set, score runs, and catch regressions when the prompt or model changes. Trigger when the user asks whether a prompt is good, wants to compare models or versions, sees inconsistent agent output, or is about to put an agent in front of customers.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

Deploys swarms of sub-agents for massive parallel data processing tasks. Unlike agent-army (which is for code changes), this is for DATA tasks -- processing 1000 documents, analyzing datasets, bulk content generation. Configurable swarm size, task distribution, result aggregation, progress tracking, and error recovery.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

Designs and deploys custom agent teams for specific business workflows. Interactive discovery of business processes, then generates complete team configurations with specialized agent roles, tool access, communication protocols, and handoff rules.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

Agent-to-Agent (A2A) communication protocol. Connect two or more Claude agents that pass messages, share context, delegate tasks, and collaborate. Implements structured handoffs, shared memory, and multi-agent conversations.

中文简介正在准备中。当前显示原文说明。

OneWave-AI/claude-skills3612026年10月20日 更新

查看 OneWave-AI 的全部技能

报告此技能的问题