agentleFS
Sign inSign up

AGENTS.md examples from real projects

How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.

Best matchesWorked for most · soon
licorsyAGENTS.md

ai-assisted-sdd-template

licorsy/ai-assisted-sdd-template/AGENTS.md

Tool-neutral agent entry point for this repository, equivalent to CLAUDE.md, for coding tools that read AGENTS.md by convention: the always-on confirmation rule, the documentation metadata rule, and pointers to the documents that own the full text of each.

02mo agoDiscuss
Twodragon0AGENTS.md

claudesec

Twodragon0/claudesec/docs/AGENTS.md

Agent guidance for ClaudeSec documentation changes and quality gates

164mo agoDiscuss
thunder-idAGENTS.md

thunderid

thunder-id/thunderid/docs/AGENTS.md

AI agents should use this file when creating and reviewing documentation content for ThunderID. It points to the docs skill that scaffolds, writes, and reviews docs, rather than restating its rules here.

5855mo agoDiscuss
yamadashyAGENTS.md

pdfvision

yamadashy/pdfvision/AGENTS.md

Core project guidelines for the pdfvision codebase. Apply these rules when working on any code, documentation, or configuration files within the pdfvision project.

2323d agoDiscuss
johnfkoo951AGENTS.md

cmds-llm-wiki

johnfkoo951/cmds-llm-wiki/AGENTS.md

Schema and harness document for the CMDS LLM Wiki vault. Defines the 3-layer architecture (Raw Sources / Wiki / Schema), ingest-query-lint operations, file conventions, and frontmatter standards. This is the single source of truth for LLM behavior in this vault.

9035d agoDiscuss
gtappsAGENTS.md

skills / references

gtapps/skills/skills/wrap/references/agents.md

Close an operator session on disk. Classify learnings verified vs idea and route them (personal memory, or the destination the project's instructions name), run the project's mandated close (proposal-act, GUEST_REPORT) and show its output, refresh the final-plan-check gauntlet with any question that bounced a plan this session, write the loose-ends line, give the plan file its issue name, print a five-line close. Trigger on "/wrap", "save learnings", "save this as a learning", "wrap this up", "wrap up", "close the session", and proactively before ending any session that produced a verdict, a plan, or a measurement. Do NOT use to commit or push (commit, commit-push), to open a PR (open-pr), or to write a handoff document for another agent (handoff).

212d agoDiscuss
obraAGENTS.md

superpowers

obra/superpowers/AGENTS.md

what produced a change in order to weigh it: agent-generated content reasoned from documentation is held to a different bar than work grounded in a real session. Contributions that

292k3mo agoDiscuss
NousResearchAGENTS.md

hermes-agent / plugins

NousResearch/hermes-agent/plugins/AGENTS.md

VERSION`, a manifest-wide native `api:` match, or version literals on unrelated payloads. Documented surfaces stay additive: - Hook payload data is added as **keyword fields**; callbacks are signature-inspected

250k6mo agoDiscuss
affaan-mAGENTS.md

ECC

affaan-m/ECC/AGENTS.md

Playwright testing | Critical user flows | | refactor-cleaner | Dead code cleanup | Code maintenance | | doc-updater | Documentation and codemaps | Updating docs | | cpp-reviewer | C/C++ code review | C and C++ projects | | cpp-build

246k30d agoDiscuss

About AGENTS.md

What is AGENTS.md?

An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.

Where does it go?

At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.

What should it contain?

Setup and test commands, code style, and the rules a new contributor would need to know.

Does Claude Code read it?

Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.