agentleFS
rstackjsrstackjs/agent-skills/AGENTS.md

agent-skills

AGENTS.md97 starsChanged 3 months agoMITView on GitHub

A collection of Agent Skills for the Rspack ecosystem (Rspack, Rsbuild, Rslib, Rstest, Rsdoctor). Skills are packaged instructions and scripts that extend agent capabilities for debugging, profiling, and development workflows. This project uses Rstack CLI as its JS toolchain: Create the Skill directory manually with the standard structure: Write the Skill content with the use cases, workflow, examples, and links to detailed references when needed. SKILL.md frontmatter field requirements: Use this for installable Skills intended for Rspack ecosystem users. -…

We split this AGENTS.md into the rules it gives an agent

13 rules. About 0 are common, 6 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read the docs linked from node_modules/rstack/docs/llms.txt when needed

    line 13
  2. Very specific:

    Run rs -h for CLI help

    line 15
  3. Distinctive:

    Create the Skill directory manually with the standard structure:

    line 48
  4. Distinctive:

    Write the Skill content with the use cases, workflow, examples, and links to detailed references when needed.

    line 58
  5. Distinctive:

    Use this for installable Skills intended for Rspack ecosystem users.

    line 68
  6. Very specific:

    Place the Skill in skills/{skill-name}/.

    line 70
  7. Very specific:

    Start SKILL.md with YAML frontmatter:

    line 75
  8. Distinctive:

    Use this for internal contribution workflow Skills intended for Rstack repository maintainers and developers, not end users.

    line 86
  9. Very specific:

    Add metadata.internal: true so the Skill is hidden from normal discovery and installation.

    line 90
  10. Very specific:

    Write code in src/index.ts:

    line 189
  11. Very specific:

    After building, the bundled files will be automatically output to the skills/my-skill/scripts/ directory.

    line 204
  12. Distinctive:

    Write tests using Rstest:

    line 208
  13. Distinctive:

    Run tests from the package directory:

    line 222

≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md

This file provides guidance to AI coding agents working with code in this repository.

Repository overview

A collection of Agent Skills for the Rspack ecosystem (Rspack, Rsbuild, Rslib, Rstest, Rsdoctor). Skills are packaged instructions and scripts that extend agent capabilities for debugging, profiling, and development workflows.

Stack

This project uses Rstack CLI as its JS toolchain:

  • Read the docs linked from node_modules/rstack/docs/llms.txt when needed
  • Online docs: https://rstack.rs/llms.txt
  • Run rs -h for CLI help

Project structure

agent-skills/
├── skills/              # Skills directory, contains all Skills
├── .agents/skills/      # Developer-facing Skills for Rstack repository maintenance
├── packages/            # Source code projects for complex scripts
├── scripts/             # Project-level configurations and tools
│   └── config/          # Shared TypeScript configuration
├── .rstack/hooks/       # Repository Git hooks
├── rstack.config.ts     # Lint, formatting, and staged-file configuration
├── pnpm-workspace.yaml  # pnpm workspace configuration
├── pnpm-lock.yaml       # Dependency lock file
├── package.json         # Project configuration file
└── README.md            # Project documentation

Directory explanations

  • skills/: Contains all Skills, each Skill is an independent folder
    • Each Skill includes SKILL.md (required), scripts/ (optional), references/ (optional), assets/ (optional)
  • .agents/skills/: Contains Skills used for Rstack repository maintenance
    • These Skills are primarily for repository developers and maintainers, not end users
  • packages/: Contains source code for complex scripts that need compilation
    • Corresponds to Skills with the same name in the skills directory
    • Compiled by Rslib through Rstack CLI and output to the corresponding skills/{skill-name}/scripts/ directory
  • scripts/config/: Contains project-level TypeScript configuration
    • tsconfig.json: TypeScript base configuration

Creating a new skill

Create the Skill directory manually with the standard structure:

{skill-name}/
├── SKILL.md          # Required: instructions + metadata
├── scripts/          # Optional: executable scripts
├── references/       # Optional: reference documentation
└── assets/           # Optional: templates, resource files

Write the Skill content with the use cases, workflow, examples, and links to detailed references when needed.

SKILL.md frontmatter field requirements:

  • name (required): unique Skill identifier, maximum 64 characters, lowercase letters/numbers/hyphens only, and must not start or end with a hyphen.
  • description (required): feature description and trigger scenarios, maximum 1024 characters. Include concrete trigger keywords, such as "when user encounters segmentation fault in Rspack".
  • metadata.internal (optional): set to true only for Rstack maintainer/developer Skills that should be hidden from normal discovery and installation.

User-facing Skills

Use this for installable Skills intended for Rspack ecosystem users.

  • Place the Skill in skills/{skill-name}/.
  • Reference bundled scripts, references, and assets with paths relative to the Skill root. Do not rely on client-specific environment variables or installation layouts.

Start SKILL.md with YAML frontmatter:

---
name: my-skill
description: Feature description and trigger scenarios, which is key for Agents to determine whether to use this Skill
---

Contribution workflow skills

Use this for internal contribution workflow Skills intended for Rstack repository maintainers and developers, not end users.

  • Place the Skill in .agents/skills/{skill-name}/.
  • Register the Skill in skills.json with "local:*" so it resolves to the repo-owned Skill source.
  • Add metadata.internal: true so the Skill is hidden from normal discovery and installation.
{
  "skills": {
    "my-internal-skill": "local:*"
  }
}
---
name: my-internal-skill
description: Internal workflow for repository maintainers
metadata:
  internal: true
---

Skills marked with metadata.internal: true are only visible and installable when INSTALL_INTERNAL_SKILLS=1 is set.

Writing skill scripts

Simple scripts

For simple scripts (such as single-file scripts), create them directly in the Skill's own scripts/ directory.

Example:

// skills/my-skill/scripts/simple.js or .agents/skills/my-skill/scripts/simple.js
console.log('Hello from simple script');

Complex scripts (Requiring Bundling)

For complex scenarios requiring dependencies, TypeScript, etc.:

1. Create a project with the same name in packages directory

packages/my-skill/
├── package.json
├── rstack.config.ts
├── tsconfig.json
└── src/
    └── index.ts

2. Configure package.json

{
  "name": "@rstackjs/my-skill",
  "version": "0.0.0",
  "type": "module",
  "scripts": {
    "build": "rs lib",
    "test": "rs test"
  },
  "devDependencies": {
    "@rstackjs/config": "workspace:*",
    "rstack": "^0.7.2"
  },
  "dependencies": {
    // Your dependencies
  }
}

3. Configure rstack.config.ts

import { basename, join } from 'node:path';
import { define } from 'rstack';

const pkgName = basename(import.meta.dirname);

define.lib({
  syntax: 'es2023',
  dts: false,
  output: {
    distPath: join(import.meta.dirname, `../../skills/${pkgName}/scripts`),
  },
});

4. Configure tsconfig.json

{
  "extends": "@rstackjs/config/tsconfig",
  "compilerOptions": {},
  "include": ["src"]
}

5. Write source code

Write code in src/index.ts:

export function myFunction() {
  // Your logic
}

6. Build script

cd packages/my-skill
pnpm build

After building, the bundled files will be automatically output to the skills/my-skill/scripts/ directory.

Testing scripts

Write tests using Rstest:

// packages/my-skill/src/index.test.ts
import { describe, it, expect } from 'rstack/test';
import { myFunction } from './index';

describe('myFunction', () => {
  it('should work correctly', () => {
    expect(myFunction()).toBe(expected);
  });
});

Run tests from the package directory:

pnpm test

Using skills

Install a specific Skill:

npx skills add rstackjs/agent-skills --skill my-skill

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.