agentleFS
javalenciacaijavalenciacai/develop-skills/AGENTS.md

develop-skills

AGENTS.md0 starsChanged 9 months agoMITView on GitHub

A comprehensive collection of AI agent skills organized as a virtual development team. These skills provide structured knowledge and workflows for software development tasks across all domains: product management, quality assurance, technical development, infrastructure, and data/AI. Each skill follows the standardized Agent Skills format: When working with this repository:

The file

Show its rules and text

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

17 rules. About 0 are common, 7 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Entry Point Pattern: All tasks flow through the CTO orchestrator

    line 12
  2. Distinctive:

    When working with this repository:

    line 131
  3. Very specific:

    Entry Point: Always start with the cto skill

    line 133
  4. Very specific:

    SKILL.md: Always uppercase, always this exact filename

    line 232
  5. Very specific:

    Context Efficiency: Keep SKILL.md focused, reference external docs when needed

    line 241
  6. Very specific:

    Create a new directory in .agents/skills/

    line 313
  7. Very specific:

    Add SKILL.md with proper YAML frontmatter

    line 314
  8. Distinctive:

    Update this AGENTS.md if needed

    line 316
  9. Distinctive:

    Test the skill with your agent

    line 317
  10. Distinctive:

    Update version info if significant changes

    line 322
  11. Distinctive:

    Test with multiple agents

    line 323
  12. Distinctive:

    Verify YAML frontmatter is valid

    line 330
  13. Very specific:

    Ensure name and description fields are present

    line 331
  14. Very specific:

    Check file is named exactly SKILL.md (uppercase)

    line 332
  15. Very specific:

    Claude Code: Ensure skills are in ~/.claude/skills/ or .claude/skills/

    line 343
  16. Very specific:

    Cursor: Check .cursor/skills/ directory

    line 344
  17. Very specific:

    GitHub Copilot: Verify .agents/skills/ path

    line 345

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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 (Claude Code, Cursor, GitHub Copilot, etc.) when working with code in this repository.

Repository Overview

A comprehensive collection of AI agent skills organized as a virtual development team. These skills provide structured knowledge and workflows for software development tasks across all domains: product management, quality assurance, technical development, infrastructure, and data/AI.

Key Features

  • Hierarchical Organization: Skills are organized in a clear organizational structure
  • Entry Point Pattern: All tasks flow through the CTO orchestrator
  • Specialized Domains: Each skill has clear domain boundaries
  • Cross-Agent Compatibility: Works with 35+ AI coding agents

Skill Architecture

Organizational Structure

CTO (Entry point)
├── PM (Product)
│   ├── PO (Requirements)
│   └── UX (Design)
├── QAL (Quality)
│   ├── QA (Testing)
│   └── SecOps (Security)
├── TL (Technical)
│   ├── Architect (Architecture)
│   └── Dev (Development)
├── InfraLead (Infrastructure)
│   ├── DevOps (CI/CD)
│   └── DBA (Databases)
└── DataLead (Data/AI)
    ├── DataEng (Pipelines)
    └── AIEng (AI/ML Models)

Directory Structure

.agents/
└── skills/
    ├── cto/
    │   └── SKILL.md
    ├── pm/
    │   └── SKILL.md
    ├── po/
    │   └── SKILL.md
    ├── ux/
    │   └── SKILL.md
    ├── qal/
    │   └── SKILL.md
    ├── qa/
    │   └── SKILL.md
    ├── secops/
    │   └── SKILL.md
    ├── tl/
    │   └── SKILL.md
    ├── architect/
    │   └── SKILL.md
    ├── dev/
    │   └── SKILL.md
    ├── infralead/
    │   └── SKILL.md
    ├── devops/
    │   └── SKILL.md
    ├── dba/
    │   └── SKILL.md
    ├── datalead/
    │   └── SKILL.md
    ├── dataeng/
    │   └── SKILL.md
    ├── aieng/
    │   └── SKILL.md
    ├── find-skills/
    │   └── SKILL.md
    └── skill-creator/
        └── SKILL.md

Skill Format

Each skill follows the standardized Agent Skills format:

SKILL.md Structure

---
name: skill-name
description: Brief description including when to use this skill
---

# Skill Title

Detailed instructions for the agent...

## When to Use

Specific scenarios where this skill should be activated.

## Responsibilities

What this skill handles.

## Workflow

Step-by-step process.

Required Fields

  • name: Unique identifier (lowercase, hyphens allowed)
  • description: Brief explanation including trigger conditions

Optional Metadata

---
name: skill-name
description: Description text
metadata:
  internal: true  # Hide from normal discovery
---

Working with These Skills

For AI Agents

When working with this repository:

  1. Entry Point: Always start with the cto skill
  2. Task Analysis: CTO analyzes the task and delegates to appropriate orchestrators
  3. Specialization: Orchestrators delegate to specialized skills
  4. Coordination: Skills can reference each other for complex workflows

Delegation Pattern

User Request
    ↓
CTO analyzes domain
    ↓
Delegates to orchestrator (PM/QAL/TL/InfraLead/DataLead)
    ↓
Orchestrator delegates to specialist (PO/UX/QA/SecOps/etc.)
    ↓
Specialist executes with domain expertise

Example Workflows

Product Feature Development

"Design and implement a new authentication feature"
    ↓
CTO → PM (product strategy)
    ↓
PM → PO (requirements) + UX (design)
    ↓
CTO → TL (technical implementation)
    ↓
TL → Architect (architecture) + Dev (code)
    ↓
CTO → QAL (quality validation)
    ↓
QAL → QA (testing) + SecOps (security audit)

Infrastructure Setup

"Set up CI/CD with Docker"
    ↓
CTO → InfraLead (infrastructure)
    ↓
InfraLead → DevOps (CI/CD setup)

AI/ML Project

"Build a recommendation engine"
    ↓
CTO → DataLead (data/AI)
    ↓
DataLead → AIEng (model) + DataEng (pipeline)

Installation

For End Users

# Install all skills
npx skills add <owner>/develop-skills

# Install specific skill
npx skills add <owner>/develop-skills --skill cto

# Install specific domain
npx skills add <owner>/develop-skills --skill tl --skill architect --skill dev

For Agent Configuration

Most agents automatically discover skills in:

  • .agents/skills/ (project-level)
  • ~/.agents/skills/ (global)

Some agents use different paths:

  • Claude Code: .claude/skills/
  • Cursor: .cursor/skills/
  • GitHub Copilot: .agents/skills/

The npx skills add command handles this automatically.

Creating New Skills

Quick Start

# Create new skill template
npx skills init my-new-skill

Naming Conventions

  • Skill directory: kebab-case (e.g., data-pipeline, api-design)
  • SKILL.md: Always uppercase, always this exact filename
  • Skill name: lowercase with hyphens in YAML frontmatter

Best Practices

  1. Clear Domain: Each skill should have a clear, specific purpose
  2. Trigger Phrases: Include specific trigger phrases in description
  3. Workflow Steps: Provide structured, numbered workflows
  4. Cross-References: Link to related skills when appropriate
  5. Context Efficiency: Keep SKILL.md focused, reference external docs when needed

Example Skill Template

---
name: my-skill
description: Brief description. Use when: (1) Scenario one, (2) Scenario two, (3) Scenario three.
---

# My Skill Name

## Role

What this skill does.

## Responsibilities

- Responsibility 1
- Responsibility 2
- Responsibility 3

## When to Use

Detailed scenarios where this skill should activate.

## Workflow

1. Step one
2. Step two
3. Step three

## Best Practices

- Practice 1
- Practice 2

## Related Skills

- **Related Skill 1**: When to use
- **Related Skill 2**: When to use

Compatibility

These skills are compatible with all agents supporting the Agent Skills specification:

  • Claude Code
  • GitHub Copilot
  • Cursor
  • Windsurf
  • Cline
  • Goose
  • Roo Code
  • And 30+ more

See skills.sh for the complete list.

Skill Discovery

The npx skills add CLI automatically discovers skills in these locations:

  • .agents/skills/
  • skills/
  • .claude/skills/
  • .cursor/skills/
  • Root directory (if SKILL.md exists)

Contributing

Adding a New Skill

  1. Create a new directory in .agents/skills/
  2. Add SKILL.md with proper YAML frontmatter
  3. Define clear use cases and workflows
  4. Update this AGENTS.md if needed
  5. Test the skill with your agent

Updating Existing Skills

  1. Maintain backward compatibility
  2. Update version info if significant changes
  3. Test with multiple agents
  4. Document breaking changes

Troubleshooting

Skill Not Loading

  • Verify YAML frontmatter is valid
  • Ensure name and description fields are present
  • Check file is named exactly SKILL.md (uppercase)
  • Confirm directory structure matches expected paths

Skills Conflict

  • Skills should have unique, non-overlapping domains
  • Orchestrator skills handle delegation
  • Specialist skills focus on execution

Agent-Specific Issues

  • Claude Code: Ensure skills are in ~/.claude/skills/ or .claude/skills/
  • Cursor: Check .cursor/skills/ directory
  • GitHub Copilot: Verify .agents/skills/ path

Meta Skills

find-skills

Helps discover and install additional agent skills from the ecosystem.

Use when: "Find a skill for...", "Is there a skill that...", "Install skills for..."

skill-creator

Guide for creating new skills with proper structure and format.

Use when: "Create a new skill", "How do I make a skill", "Design a skill for..."

License

MIT

Learn More

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.