Community Reference

AI Agent
Instruction Files

A cross-tool reference for AGENTS.md, CLAUDE.md, copilot-instructions.md, GEMINI.md, and every other file that tells your AI coding agents how to work.

↓ Compatibility Matrix File Reference →
10+
Instruction file types
6
Major tools covered
60k+
Repos using AGENTS.md
2025
Standard formalized

Tool Compatibility Matrix

Which files each AI coding tool reads natively. ✓ = native support, ~ = via symlink, — = not supported.

File Claude Code GitHub Copilot OpenAI Codex Gemini CLI Cursor Windsurf Scope
AGENTS.md
repo root or subdirs
~ Universal
CLAUDE.md
root / ~/.claude / subdirs
~ ~ Claude
.github/copilot-instructions.md
.github/ folder
Copilot only
*.instructions.md
.github/instructions/
Copilot only
*.agent.md
.github/agents/
Copilot only
*.prompt.md
.github/prompts/
Copilot only
GEMINI.md
repo root
~ ~ Gemini
SKILL.md
.github/skills/ · .claude/skills/ · .windsurf/skills/ · ~/.copilot/skills/ · Microsoft Agent Framework (✓ native)
Shared
.claude/agents/*.md
project or ~/.claude/agents/
Claude only
.claude/commands/*.md
project or ~/.claude/commands/
Claude only
.cursor/rules/*.mdc
.cursor/rules/
Cursor only
openai-codex/setup.sh
repo root
Codex only

Per-File Reference

What each file does, where it lives, and a starter template you can copy.

AGENTS.md
/ or any subdirectory

The open standard. A README for agents. Supported by Copilot, Codex, Gemini CLI, Cursor, Windsurf, Zed, and more. Most-nested file takes precedence. Plain Markdown — no required schema.

universal hierarchical open standard
Starter template
# AGENTS.md
## Project
Next.js 14 app with TypeScript + Postgres.

## Commands
- Install: `pnpm install`
- Dev: `pnpm dev`
- Test: `pnpm test`
- Lint: `pnpm lint:fix`

## Standards
- Named exports only
- Zod for all validation
- Tests required on all PRs
CLAUDE.md
root / ~/.claude / subdirs

Claude Code's primary memory file. Loaded every session. Hierarchy: enterprise → user global (~/.claude/CLAUDE.md) → project root → subdirectory. More specific overrides broader. Target: under 300 lines.

claude code auto-loaded hierarchical
Key sections
# CLAUDE.md
# Keep under 300 lines. Point to
# other files for specific topics.

## What this project is
One sentence overview.

## How to run things
Exact commands only. No vague hints.

## What NOT to do
NEVER modify /migrations directly.

## More context
See agent_docs/ for detailed guides.
copilot-instructions.md
.github/copilot-instructions.md

Global instructions for all GitHub Copilot chat and the coding agent. Applied automatically to every request in the repo. One file, repo-wide scope only.

github copilot repo-wide chat + agent
Starter template
# Applied to all Copilot requests

We use Bazel for Java deps, not Maven.
JS uses double quotes + tab indentation.
We use Jira for issue tracking.
Run `npm test` before any PR.
All API routes live in /src/api/.
*.instructions.md
.github/instructions/

Path-specific instructions for Copilot. Uses YAML frontmatter with glob patterns to apply rules only to matching files. Multiple files, granular control. Use excludeAgent to exclude specific Copilot agents.

github copilot path-scoped frontmatter
Example: models.instructions.md
---
applyTo: "src/models/**/*.ts"
excludeAgent: "code-review"
---

All models extend BaseModel.
Use Zod schemas for validation.
Export both type and schema.
GEMINI.md
repo root

Gemini CLI's equivalent of CLAUDE.md. Also read by GitHub Copilot when Gemini model is active. Same structure as AGENTS.md — plain Markdown, no required schema.

gemini cli copilot gemini model
Structure mirrors AGENTS.md
# GEMINI.md
# Same structure as AGENTS.md.
# Symlink if content is identical:

ln -s AGENTS.md GEMINI.md

# Only create separately if you need
# Gemini-specific instructions.
SKILL.md
.github/skills/[name]/ · .claude/skills/[name]/ · ~/.copilot/skills/ · ~/.claude/skills/

Reusable skill bundles loaded on demand — not at every session start. A skill is a folder containing SKILL.md plus any helper scripts or resource files. Portable across Claude Code and GitHub Copilot (Copilot CLI, coding agent, VS Code agent mode). Each tool uses the YAML frontmatter description field to decide when to load the skill. Copilot also supports manual invocation via /skill-name in chat. Project skills are repo-scoped; personal skills (in ~/.copilot/skills/ or ~/.claude/skills/) apply across all projects.

claude code github copilot on-demand portable
Structure: .github/skills/api-tests/SKILL.md
---
name: api-tests
description: Write and run API integration tests.
  Use when asked to create or fix API tests.
---

# Step-by-step instructions go here.
# Include helper scripts and examples
# in the same folder alongside SKILL.md.
.claude/agents/*.md
project or ~/.claude/agents/

Custom subagents for Claude Code with isolated context. Defined via YAML frontmatter. Project agents (.claude/agents/) are version-controlled and team-shared. User agents (~/.claude/agents/) are personal and available in all projects.

claude code subagents frontmatter
Example: .claude/agents/reviewer.md
---
name: code-reviewer
description: Reviews PRs for bugs
model: claude-sonnet-4
tools: [read_file, bash]
---

You are a principal-level code reviewer.
Focus on bugs, security, and performance.
Be specific — cite file and line number.
.cursor/rules/*.mdc
.cursor/rules/

Cursor's rule system with four loading modes: Always (every context), Auto Attached (glob-matched files), Agent Requested (AI decides), Manual (explicit only). More granular than AGENTS.md for Cursor power users.

cursor conditional loading 4 modes
Example with frontmatter
---
description: DynamoDB model patterns
globs: "src/models/**/*.py"
alwaysApply: false
---

Use PynamoDB for all DynamoDB models.
Hash key must be named `pk`, range `sk`.

Projects Worth Knowing

Tools and collections that extend, distribute, or orchestrate AI agent instruction files. Verified details only — compatibility marked explicitly.

APM — Agent Package Manager
Microsoft · @danielmeppiel
github.com/microsoft/apm ↗

Think package.json but for AI agent configuration. APM introduces an apm.yml manifest that declares the skills, prompts, instructions, and agents your project needs. Run apm install and every developer gets an identical agent setup. Resolves transitive dependencies, compiles to AGENTS.md and CLAUDE.md automatically, and installs from GitHub, GitHub Enterprise, and Azure DevOps — no registry required.

Tool compatibility
✓ GitHub Copilot ✓ Claude / Claude Code ✓ Cursor ✓ OpenAI Codex ✓ Gemini
License: MIT Language: Python Install: pip install apm-cli

An AI team framework built on top of GitHub Copilot. You describe what you're building and Squad proposes a named team of specialist agents — each with a defined role, its own memory, and shared decision logs. Agents persist across sessions as files in your repo, learning your codebase over time. Integrates with GitHub Issues via labels: tag an issue squad and the lead agent triages and assigns it automatically. Can include @copilot as a team member for autonomous coding tasks. TypeScript SDK-first mode available via squad.config.ts. Featured on the GitHub Blog (March 2026).

Tool compatibility
✓ GitHub Copilot Not supported — Cursor Not supported — Claude Code Not supported — Codex
Install: npm install -g @bradygaster/squad-cli Docs: bradygaster.github.io/squad
anthropics/skills
@anthropics (Anthropic)
github.com/anthropics/skills ↗

Anthropic's official skill library — the canonical reference for the SKILL.md format. Contains 50+ production-quality skills across document processing (Word, PDF, PowerPoint, Excel), development tools, design, and enterprise workflows. The docx, pdf, pptx, and xlsx skills are the same ones powering Claude's document capabilities in production. Installable as a Claude Code plugin via /plugin install.

Tool compatibility
✓ Claude / Claude Code ✓ GitHub Copilot ✓ OpenAI Codex CLI Not supported — Cursor (natively)
License: Apache 2.0 (most skills) Install: /plugin install document-skills@anthropic-agent-skills

GitHub's official community-contributed collection of Copilot customizations. Includes instructions files, agent personas, hooks (event-driven automations triggered by Copilot coding agent sessions), agentic workflows for GitHub Actions, plugins, and skills. A practical starting point for teams looking to adopt instruction files — most entries are production-tested and cover a wide range of languages, frameworks, and roles.

Tool compatibility
✓ GitHub Copilot ✓ Claude Code (skills) Not supported — Codex Not supported — Cursor
License: MIT Community PRs welcome
agentsmd/agents.md
@agentsmd — OpenAI, Google, Cursor, Factory, Sourcegraph
github.com/agentsmd/agents.md ↗

The canonical specification repository for the AGENTS.md open standard. Released by OpenAI in August 2025 and donated to the Linux Foundation / Agentic AI Foundation in December 2025. Includes the formal spec, examples, and a companion site at agents.md. This is the authoritative source for how the standard should behave across tools — read this before writing your first AGENTS.md.

Standard adopted by
✓ GitHub Copilot ✓ OpenAI Codex ✓ Gemini CLI ✓ Cursor ✓ Windsurf ✓ Zed ~ Claude Code (via symlink)
Site: agents.md Governed by: Linux Foundation AAIF
GitHub Copilot SDK
GitHub · @github
github.com/github/copilot-sdk ↗

Embeds the same agentic execution loop that powers Copilot CLI directly into your application. Instead of building your own orchestration layer, you define agent behavior — tools, instructions, permissions — and Copilot handles planning, tool invocation, file edits, multi-turn memory, and model routing. Supports BYOK (Bring Your Own Key) so you can substitute OpenAI, Azure AI Foundry, or Anthropic models without a Copilot subscription. Available in Python, TypeScript, Go, and .NET. Launched January 2026; currently in technical preview.

Tool compatibility
✓ GitHub Copilot ✓ BYOK (OpenAI / Azure / Anthropic) Not supported — Cursor Not supported — Claude Code
Status: Technical Preview Languages: Python, TypeScript, Go, .NET Install: pip install copilot-sdk / npm install @github/copilot-sdk

A portable, open-source trust policy and guardrail specification for agentic systems. Defines agent behavior boundaries — what actions are permitted, denied, or require human approval — in a single bouncer.md file that any agent runtime can read. Designed to be composable alongside AGENTS.md and CLAUDE.md: instruction files tell agents what to do, bouncer-md defines the guardrails within which they operate. Currently v0.5 with a full repo structure, test suite, and examples. Maintained independently of any tool vendor to preserve community credibility.

Compatibility
✓ Runtime-agnostic spec ✓ Claude Code ✓ GitHub Copilot ✓ OpenAI Codex
License: MIT Status: v0.5 Docs: bouncer-md.github.io


Tips That Actually Matter

Distilled from real-world usage and analysis of thousands of repos.

🪶
Keep it short

Every line in your memory file competes with actual work for context window. Target under 300 lines for CLAUDE.md. Under 200 lines before splitting into subdirectory files for AGENTS.md.

🔗
Point, don't paste

Reference other files instead of inlining content. Keep task-specific docs in agent_docs/ and list them in your root file. Agents read what's relevant; unused context is wasted tokens.

🚫
Never use for code style

Don't put linting rules or formatting preferences in instruction files. Use ESLint, Prettier, or your formatter. LLMs are slow and expensive compared to deterministic tools. Reserve instructions for what only a human can explain.

Commands, not descriptions

Write the exact command (pnpm test --run --no-color), not a description of what it does. Include flags. Agents use these literally — vague wording causes hallucinated flags.

📁
Use hierarchy for monorepos

Place additional AGENTS.md files in subdirectories. The nearest file in the directory tree wins. OpenAI's own repo uses 88 AGENTS.md files. Don't try to cover everything from the root.

🔄
Treat it like code

Update instruction files in the same PR as the process change. Review them in code review. Audit quarterly. Outdated instructions actively hurt agent performance — worse than no instructions at all.

🧱
Set clear boundaries

Tell agents what NOT to do as explicitly as what to do. Include a "never do" section. Tell agents when to ask for human review rather than proceeding. Agents that know their limits are more reliable.

🔑
Never commit secrets

These files go into your system prompt and are version-controlled. Never include API keys, passwords, connection strings, or security vulnerability details. Treat the file as public documentation.


What's Changed

Major changes to the ecosystem. This space moves fast — bookmark and check back.

Apr 3, 2026
New
Microsoft Agent Framework 1.0 GA The unified successor to Semantic Kernel and AutoGen ships production-ready for .NET and Python, with stable APIs and a long-term support commitment. Includes native SKILL.md support via FileAgentSkillsProvider — same progressive disclosure pattern as Claude Code and Copilot. AGENTS.md read natively for project instructions. Semantic Kernel and AutoGen enter maintenance mode. Announcement ↗ GitHub ↗
Mar 11, 2026
New
Copilot for JetBrains adds AGENTS.md and CLAUDE.md support The March JetBrains update adds native support for AGENTS.md and CLAUDE.md instruction files, including nested file discovery and auto-generation of an initial AGENTS.md. A /memory slash command provides quick access to manage instruction files. Copilot auto-discovers and loads instruction files into context during agent workflows. Changelog ↗
Mar 30, 2026
New
bouncer-md v0.5 launches A portable, open-source trust policy and guardrail specification for agentic systems. Defines permitted, denied, and approval-gated actions in a single bouncer.md file composable alongside existing instruction files. Runtime-agnostic by design, maintained independently of any tool vendor. Ships with a full repo structure, test suite, and examples. GitHub ↗ Docs ↗
Mar 18, 2026
New
Next.js 16.2 auto-generates AGENTS.md and CLAUDE.md create-next-app now scaffolds both files by default. AGENTS.md directs agents to version-matched docs bundled inside node_modules/next/dist/docs/ — no stale training data. CLAUDE.md uses the @ import syntax to reference AGENTS.md without duplication. Vercel reports 100% pass rate on internal evals with this context in place. Blog Post ↗ Docs ↗
Mar 9, 2026
New
Windsurf adds SKILL.md support Windsurf's March 9 release adds loading of SKILL.md files from .windsurf/skills/, expanding the compatibility matrix. Also adds system-level skill definitions via MDM-managed configs for Enterprise deployments. Changelog ↗
Feb 12, 2026
Research
ETH Zurich study questions AGENTS.md effectiveness Researchers tested four coding agents across 138 real-world tasks. LLM-generated context files reduced task success rates and raised inference costs ~20%. Human-written files yielded a marginal +4% improvement but also increased costs ~19%. Recommendation: keep files minimal, human-written, and focused on non-inferable details only. Paper ↗ InfoQ Coverage ↗
Jan 23, 2026
New
GitHub Copilot SDK enters technical preview Embeds Copilot's full agentic runtime (planning, tool use, file edits, memory) into any application. Available in Python, TypeScript, Go, and .NET. Supports BYOK for OpenAI, Azure AI Foundry, and Anthropic models. Announcement ↗ GitHub ↗
Aug 2025
Standard
AGENTS.md formalized as open standard Collaborative effort between OpenAI, Google, Cursor, Factory, and Sourcegraph. Now maintained under the Agentic AI Foundation / Linux Foundation. 20,000+ repos adopted. agents.md ↗ GitHub ↗
Aug 28, 2025
New
GitHub Copilot adds native AGENTS.md support Copilot coding agent now reads AGENTS.md in addition to copilot-instructions.md, CLAUDE.md, and GEMINI.md. Changelog ↗
Dec 18, 2025
New
GitHub Copilot Agent Skills launched Copilot now supports SKILL.md natively — the same format as Claude Code skills. Store project skills in .github/skills/ or .claude/skills/; personal skills in ~/.copilot/skills/. Invoke with /skill-name in chat or let Copilot auto-load based on the description frontmatter field. Changelog ↗ Docs ↗
Dec 9, 2025
Standard
AGENTS.md donated to Linux Foundation / Agentic AI Foundation OpenAI contributed AGENTS.md to the newly formed AAIF alongside Anthropic's MCP. Platinum members include AWS, Anthropic, Google, Microsoft, and OpenAI. Adoption now exceeds 60,000 open-source projects. AAIF ↗
Nov 2025
Update
Copilot adds excludeAgent property to *.instructions.md Fine-grained control over which Copilot agent (chat, code review, coding agent) uses a given instructions file. Changelog ↗
Jul 2025
New
Copilot coding agent adds *.instructions.md support Path-specific instructions via YAML frontmatter applyTo glob patterns. Stored in .github/instructions/. Docs ↗
Early 2025
New
Claude Code 2.0 released with subagents and skills Introduced .claude/agents/, .claude/commands/, and .claude/skills/ directory conventions. Skills shareable across claude.ai, Claude Code, and Claude Desktop. Claude Code Docs ↗ GitHub ↗
2024–2025
Deprecated
.cursorrules file deprecated by Cursor Replaced by the more powerful .cursor/rules/*.mdc system with four loading modes and glob-based conditional application. Cursor Rules Docs ↗