💡 摘要
评估存储库在八个关键支柱上的自主AI代理准备情况。
🎯 适合人群
🤖 AI 吐槽: “看起来很能打,但别让配置把人劝退。”
风险:Medium。建议检查:是否执行 shell/命令行指令;是否发起外网请求(SSRF/数据外发);文件读写范围与路径穿越风险。以最小权限运行,并在生产环境启用前审计代码与依赖。
readiness
Agent Readiness skill for Claude Code - evaluate repository readiness for autonomous AI agents.
Inspired by Factory AI Agent Readiness.
Installation
Via skills.sh (recommended)
npx skills add viktor-silakov/readiness
This is the easiest way to install. The skill will automatically appear in the skills.sh leaderboard based on installation count.
As Claude Code Plugin
claude plugin install readiness@https://github.com/viktor-silakov/readiness
Manual Installation
Clone to your personal skills directory:
git clone https://github.com/viktor-silakov/readiness ~/.claude/skills/readiness
Or to project-level:
git clone https://github.com/viktor-silakov/readiness .claude/skills/readiness
Via npx
npx claude-skill-readiness
This installs the skill to .claude/skills/readiness/ in your project.
Usage
In Claude Code, run:
/readiness
Or with plugin:
/readiness:readiness
What it does
The skill evaluates your repository across 8 pillars and assigns a maturity level 1-5:
8 Pillars
| Pillar | What's checked | |--------|----------------| | Style & Validation | ESLint, Prettier, Biome, formatters | | Build System | package.json scripts, tsconfig, build tools | | Testing | Jest, Vitest, Playwright, test coverage | | Documentation | README, CLAUDE.md/AGENTS.md | | Dev Environment | .nvmrc, .env.example, docker-compose | | Code Quality | TypeScript strict mode, pre-commit hooks | | Observability | Logging, error tracking | | Security | .gitignore, secrets, dependencies audit |
5 Maturity Levels
| Level | Name | Description | |-------|------|-------------| | 1 | Functional | Build works, has entry point | | 2 | Documented | README, installation instructions | | 3 | Standardized | Tests, linting, TypeScript strict | | 4 | Optimized | CI/CD, pre-commit hooks, coverage | | 5 | Autonomous | AGENTS.md, E2E tests, CODEOWNERS |
Output Example
+==============================================================+
| AGENT READINESS REPORT |
+==============================================================+
| Repository: my-project |
| Level Achieved: 4 (Optimized) |
| Overall Score: 78% |
+--------------------------------------------------------------+
PILLAR SCORES
+-------------------------------------------------------------+
| Style & Validation ████████████░░░░ 75% (3/4 criteria) |
| Build System ████████████████ 100% (4/4 criteria) |
| Testing ██████████░░░░░░ 62% (5/8 criteria) |
| ... |
+-------------------------------------------------------------+
Auto-fix
After generating the report, the skill offers to fix common issues:
- Create AGENTS.md with AI agent instructions
- Add .nvmrc with Node version
- Create CODEOWNERS file
- Add missing .gitignore entries
- Configure pre-commit hooks
CLI Commands
npx claude-skill-readiness install # Install skill (default) npx claude-skill-readiness check # Check if installed npx claude-skill-readiness update # Update skill files npx claude-skill-readiness remove # Remove skill
Requirements
- Node.js >= 18
- Claude Code CLI
License
MIT
优点
- 涵盖多个标准的全面评估。
- 自动生成带有成熟度等级的报告。
- 提供常见问题的自动修复建议。
缺点
- 需要Node.js和Claude Code CLI。
- 可能无法覆盖所有特定项目需求。
- 依赖外部工具以实现完整功能。
相关技能
免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。
版权归原作者所有 viktor-silakov.
