Agent rules

Skill mr-kelly/skills/skills/agent-rules

Align all AI coding agents to single source of truth. AGENTS.md for rules, .agents/skills/ for skills. All other files are symlinks.From its SKILL.md

Install

npx -y skills add mr-kelly/skills --skill agent-rules

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.8 KB, as published. Nobody here has run it

Agent Rules - Single Source of Truth

Ensure all AI coding agents use the same rules and skills via symlinks.

Quick Start

/agent-rules              # Align all agents
/agent-rules --check      # Verify alignment
/agent-rules --status     # Show current status

Single Source of Truth

Two sources only:

Rules: AGENTS.md (project root) - edit here
Skills: .agents/skills/ - edit here (SOURCE)

Everything else is a symlink:

CLAUDE.md → AGENTS.md
.github/copilot-instructions.md → ../AGENTS.md
.cursor/rules/main.md → ../../AGENTS.md
.gemini/GEMINI.md → ../AGENTS.md
.kiro/steering/main.md → ../../AGENTS.md
.claude/skills/ → ../.agents/skills
.github/skills/ → ../.agents/skills
.kiro/skills/ → ../.agents/skills
.codex/skills/ → ../.agents/skills

Architecture

AGENTS.md                         ← Edit rules here
.agents/skills/                   ← Edit skills here (SOURCE)

↓ All others are symlinks ↓

CLAUDE.md → AGENTS.md
.github/copilot-instructions.md → ../AGENTS.md
.cursor/rules/main.md → ../../AGENTS.md
.gemini/GEMINI.md → ../AGENTS.md
.kiro/steering/main.md → ../../AGENTS.md

.claude/skills/ → ../.agents/skills
.github/skills/ → ../.agents/skills
.kiro/skills/ → ../.agents/skills
.codex/skills/ → ../.agents/skills

Supported Agents

Agent	Skills	Instructions
OpenAI Codex	`.codex/skills/` →	`AGENTS.md`
Claude Code	`.claude/skills/` →	`CLAUDE.md` →
GitHub Copilot	`.github/skills/` →	`.github/copilot-instructions.md` →
Kiro CLI	`.kiro/skills/` →	`.kiro/steering/main.md` →
Cursor IDE	N/A	`.cursor/rules/main.md` →
Gemini CLI	N/A	`.gemini/GEMINI.md` →

All skills symlinks point to .agents/skills/

Usage

Edit Rules

vim AGENTS.md    # Edit once, all agents see it

Add Skill

mkdir -p .agents/skills/my-skill
vim .agents/skills/my-skill/SKILL.md
# All agents see it automatically

Align All Agents

/agent-rules

Critical Rules

✅ ONLY edit AGENTS.md - single source for rules
✅ ONLY edit .agents/skills/ - single source for skills
❌ NEVER edit symlinked files - they're just links
❌ NEVER copy files - use symlinks only

What About MCP Configs?

MCP (Model Context Protocol) configs are NOT symlinked because:

Each agent has different config format (.json vs .toml)
Contains agent-specific settings
May contain sensitive info (API keys)

MCP config locations:

Claude: .claude/mcp.json
Kiro: .kiro/mcp.json
Cursor: .cursor/mcp.json
Codex: .codex/config.toml

If you need to sync MCP configs: Do it manually or create a sync script.

Commands

/agent-rules              # Create all symlinks
/agent-rules --check      # Verify all symlinks valid
/agent-rules --status     # Show symlink status
/agent-rules --fix        # Fix broken symlinks

Scripts

scripts/agent-rules.sh - Main entry point
scripts/create-symlinks.sh - Create symlinks
scripts/verify-architecture.sh - Verify alignment

Troubleshooting

Symlink broken

/agent-rules --fix

Skill not found

ls .agents/skills/skill-name/  # Check source exists

Version: 4.0
Last Updated: 2026-02-20
Breaking Change: Skills source moved from .claude/skills/ to .agents/skills/

What ships with it: 4 files

9.7 KB alongside SKILL.md, 3 of them executable

scripts/

README.md1.8 KB

Gives 0 of the 12 instructions most context ai engineering skills give

Counted across 1,328 of the 2,349 authors here whose files we hold, read 2026-09-06

Dispatch a fresh subagent for each taskin 76 of 1328, across 59 files
Perform spec compliance review before code quality reviewin 44 of 1328, across 34 files
Dispatch a final code reviewer after all tasksin 38 of 1328, across 26 files
Answer subagent questions before allowing implementationin 36 of 1328, across 26 files
Use the least powerful model capable of the taskin 33 of 1328, across 26 files
Create a TodoWrite list for all tasksin 32 of 1328, across 22 files
Perform a task review after each implementationin 31 of 1328, across 24 files
Extract all tasks and context from the planin 29 of 1328, across 20 files
Provide full task text to subagentsin 28 of 1328, across 20 files
Use git worktrees for isolated workspacesin 25 of 1328, across 20 files
Specify the model explicitly when dispatching a subagentin 23 of 1328, across 18 files
Execute all tasks from the plan without stoppingin 21 of 1328, across 16 files

Said here and by no other author read

edit skills only in .agents/skills/
run agent-rules to align all agents
run agent-rules --fix to repair broken symlinks

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.