agents-generator Skill
生成符合 agents.md 标准(60k+ 开源项目使用)的完整项目级 AGENTS.md。还能为检测到的平台生成对应文件(.cursorrules、Copilot 指令、GEMINI.md)。在配置 AI agent 规则、为 Claude Code、Cursor、Copilot、OpenCode 或任何读取 AGENTS.md 的 agent 建立项目规范时使用。支持三种模式:full(完整 AGENTS.md + 配套规则文件 + 多平台输出)、minimal(单个 30 行的符合 agents.md 标准的 AGENTS.md)和 update(对比差异,仅重新生成改动部分)。可检测 monorep
安装方式:把技能目录放入 ~/.claude/skills/(Claude Code)或在 claude.ai 设置中启用;也可复制右侧安装命令一键添加。
技能指令原文(SKILL.md)
Skill: agents-generator
Generates a tailored AGENTS.md + .agents/rules/*.md for the target project — not a template with placeholders, but a living document that matches the project's real toolchain.
What you get
From a project that uses Bun + Next.js 16 + Tailwind + Vitest + Server Actions, the skill produces:
AGENTS.md
├── Setup commands: bun install, bun dev, bun run test:run, bun doctor
├── Verification Cycle: bunx tsc --noEmit → bun run lint → bun run test:run → bun doctor
├── Conventions: "Bun always. Plain TypeScript types + guards."
└── Architecture → .agents/rules/architecture.md
.agents/rules/
├── architecture.md ← ASCII diagram with real directories, exact versions
├── frontend-patterns.md ← Component rules, state locations, trust boundaries
├── server-actions.md ← downloadVideo() flow, DownloadResult type, rate limiter
├── testing.md ← "74 tests in 5 files", vitest commands, mock patterns
├── git-workflow.md ← Conventional commits, pre-commit checks
└── sdd-workflow.md ← Preflight defaults, post-apply verification
Rules NOT generated: backend.md (no NestJS), database.md (no ORM), i18n.md (hardcoded Spanish), forms.md (manual inputs), styling.md (Tailwind in frontend rules).
Activation Contract
Generate AGENTS.md + .agents/rules/*.md for the target project. Never guess — read the project's actual files first.
Mode selection
| User says | Mode | Output |
|-----------|------|--------|
| "simple AGENTS.md", "just the basics", "minimal" | Minimal | Single AGENTS.md (~30 lines, no rule files) |
| "full AGENTS.md", "with rules", "complete", or default | Full | AGENTS.md + .agents/rules/*.md |
| "update AGENTS.md", "refresh", "my stack changed" | Update | Diff existing, regenerate only what changed |
Dry-run mode
If the user asks to "preview", "show what would change", "dry-run": run all detection but do NOT write files. Show detection summary, files that would be created, skipped rules, and sample output.
Hard Rules
- Read before writing. Read
package.json, all config files, and directory structure before generating anything. - Detect package manager FIRST. Check lockfiles:
bun.lock→bun,pnpm-lock.yaml→pnpm,package-lock.json→npm,yarn.lock→yarn. NEVER default to npm. Every command uses the detected PM. - Generate only what applies. No backend rules for frontend-only. No database rules without ORM.
- Auto-format and lint generated files. Run
[format cmd]and[lint cmd]on generated files only — never the whole project. These are fast, safe operations. - Validate commands. Every command in output must exist as a script key in
package.json. - No placeholders. Scan output for
{{,TODO,add here,.... Reject if any remain. - Backup first. If files exist, copy to
.agents/backups/with timestamp.
Execution Steps
Common
git rev-parse --show-toplevel→ project root.- Detect package manager FIRST: check lockfiles.
bun.lock→bun,pnpm-lock.yaml→pnpm,package-lock.json→npm,yarn.lock→yarn. Never default to npm. - Read
package.json(scripts, deps, workspaces). Save scripts for validation. - Read config files and explore directory structure.
- Select mode (ask if ambiguous).
Full mode
- Read
assets/agents-full.md— this is the AGENTS.md structure with all sections and filling rules. - Read project files and fill every placeholder with real data. Never use generic text.
- Generate
AGENTS.mdat project root. Wrap content in/. - For each applicable rule category, read the corresponding template from
assets/and generate the rule file in.agents/rules/. - If Claude detected (
.claude/orCLAUDE.md): generate thinCLAUDE.mdfromassets/claude.md. - If platform files detected: generate from
assets/platform.md.
Minimal mode
- Read
assets/agents-minimal.md— 30-line agents.md standard format. - Generate single
AGENTS.md.
Update mode
- Backup existing files.
- Re-detect project state.
- Diff old vs new. Regenerate only changed categories.
Post-generation
- Run
[format cmd]on the generated files only. - Run
[lint cmd]on the generated files only. - Scan for
{{,TODO,.... Fix any found. - Verify all commands exist in package.json scripts.
- If AGENTS.md > 300 lines, warn. If > 500, move content to rule files.
- Summarize all changes using conventional commit format before declaring done.
- Report: what was detected, generated, skipped, and confidence score.
Output Contract
Return:
- Mode used and why
- Files created/modified
- Detection summary (all categories)
- Rules generated and skipped (with reason)
- Confidence score
References
| Priority | File | Purpose |
|----------|------|---------|
| Required | assets/agents-full.md | Full AGENTS.md template with all 25+ sections and filling rules |
| Required | assets/agents-minimal.md | 30-line agents.md standard template |
| Full mode | assets/architecture.md | Architecture rules template |
| Full mode | assets/frontend-patterns.md | Frontend patterns template |
| Full mode | assets/server-actions.md | Server actions / backend template |
| Full mode | assets/testing.md | Testing strategy template |
| Full mode | assets/git-workflow.md | Git workflow template |
| Full mode | assets/sdd-workflow.md | SDD workflow template |
| Full mode | assets/styling.md | Styling rules template |
| Full mode | assets/forms.md | Form patterns template |
| Full mode | assets/database.md | Database rules template |
| Full mode | assets/i18n.md | i18n rules template |
| Full mode | assets/backend.md | Backend/NestJS template |
| Conditional | assets/claude.md | CLAUDE.md — only if Claude detected |
| Conditional | assets/platform.md | Multi-platform files |
| Conditional | assets/agents-nested.md | Monorepo nested AGENTS.md |
| Reference | references/decision-matrix.md | Full detection logic and edge cases |
| Reference | references/example-output/README.md | Quality benchmark |
| Reference | references/template-filling-guide.md | Placeholder filling rules |