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:
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,542 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository34 other files it gives its agents.
- .agents/agents.mdAGENTS.md
- .agents/skills/aieng/agents.mdAGENTS.md
- .agents/skills/architect/agents.mdAGENTS.md
- .agents/skills/cto/agents.mdAGENTS.md
- .agents/skills/dataeng/agents.mdAGENTS.md
- .agents/skills/datalead/agents.mdAGENTS.md
- .agents/skills/dba/agents.mdAGENTS.md
- .agents/skills/dev/agents.mdAGENTS.md
- .agents/skills/devops/agents.mdAGENTS.md
- .agents/skills/infralead/agents.mdAGENTS.md
- .agents/skills/pm/agents.mdAGENTS.md
- .agents/skills/po/agents.mdAGENTS.md
- .agents/skills/qa/agents.mdAGENTS.md
- .agents/skills/qal/agents.mdAGENTS.md
- .agents/skills/secops/agents.mdAGENTS.md
- .agents/skills/tl/agents.mdAGENTS.md
- .agents/skills/ux/agents.mdAGENTS.md
- aiengSkill
- architectSkill
- ctoSkill
- dataengSkill
- dataleadSkill
- dbaSkill
- devopsSkill
- devSkill
- infraleadSkill
- pmSkill
- poSkill
- qalSkill
- qaSkill
- secopsSkill
- tlSkill
- uxSkill
- orchestratorSkill
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
- Distinctive: line 12
Entry Point Pattern: All tasks flow through the CTO orchestrator
- Distinctive: line 131
When working with this repository:
- Very specific: line 133
Entry Point: Always start with the cto skill
- Very specific: line 232
SKILL.md: Always uppercase, always this exact filename
- Very specific: line 241
Context Efficiency: Keep SKILL.md focused, reference external docs when needed
- Very specific: line 313
Create a new directory in .agents/skills/
- Very specific: line 314
Add SKILL.md with proper YAML frontmatter
- Distinctive: line 316
Update this AGENTS.md if needed
- Distinctive: line 317
Test the skill with your agent
- Distinctive: line 322
Update version info if significant changes
- Distinctive: line 323
Test with multiple agents
- Distinctive: line 330
Verify YAML frontmatter is valid
- Very specific: line 331
Ensure name and description fields are present
- Very specific: line 332
Check file is named exactly SKILL.md (uppercase)
- Very specific: line 343
Claude Code: Ensure skills are in ~/.claude/skills/ or .claude/skills/
- Very specific: line 344
Cursor: Check .cursor/skills/ directory
- Very specific: line 345
GitHub Copilot: Verify .agents/skills/ path
≈ “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.
Files
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.mdSkill 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:
- Entry Point: Always start with the
ctoskill - Task Analysis: CTO analyzes the task and delegates to appropriate orchestrators
- Specialization: Orchestrators delegate to specialized skills
- 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 expertiseExample 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 devFor 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-skillNaming 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
- Clear Domain: Each skill should have a clear, specific purpose
- Trigger Phrases: Include specific trigger phrases in description
- Workflow Steps: Provide structured, numbered workflows
- Cross-References: Link to related skills when appropriate
- 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 useCompatibility
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
- Create a new directory in
.agents/skills/ - Add
SKILL.mdwith proper YAML frontmatter - Define clear use cases and workflows
- Update this AGENTS.md if needed
- Test the skill with your agent
Updating Existing Skills
- Maintain backward compatibility
- Update version info if significant changes
- Test with multiple agents
- Document breaking changes
Troubleshooting
Skill Not Loading
- Verify YAML frontmatter is valid
- Ensure
nameanddescriptionfields 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
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
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.

