forgeos
AGENTS.md2 starsChanged 2 months agoMITView on GitHub
Entry point for Codex and any other agent that reads AGENTS.md. It holds no rules of its own. Every rule lives in the operating contract, which Claude Code loads through CLAUDE.md, so both tools obey one source of truth. .ai/contract/core.md, in full, before anything else. It is authoritative: the discovery gate, instruction priority, context policy, the non-negotiable rules, execution principles, lifecycle, validation, safety, reporting, and the reference map. Section 0 is the discovery gate and applies before any other section.
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 31,774 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 repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- codebase-navigationSkill
- debuggingSkill
- documentation-updateSkill
- refactoringSkill
- test-generationSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 4
Every rule lives in the operating contract, which Claude Code loads through CLAUDE.md, so both tools obey one source of truth.
≈ “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
Entry point for Codex and any other agent that reads AGENTS.md. It holds no rules of its own.
Every rule lives in the operating contract, which Claude Code loads through CLAUDE.md, so both
tools obey one source of truth.
Read this first
.ai/contract/core.md, in full, before anything else. It is authoritative: the discovery gate,
instruction priority, context policy, the non-negotiable rules, execution principles, lifecycle,
validation, safety, reporting, and the reference map.
Section 0 is the discovery gate and applies before any other section.
Then, and only then:
.ai/context/project.md— stable project facts.ai/context/constraints.md— hard constraints- The active task in
.ai/tasks/active/
Section 2 of the contract says what else exists and which trigger loads it. Nothing is loaded by default, and nothing on this page substitutes for reading it.
Where things live
- Engineering rules:
.ai/rules/· Procedures:.ai/workflows/· Techniques:.ai/skills/ - Roles:
.ai/agents/· Project types:.ai/profiles/· Filled records:examples/
Validation
powershell -NoProfile -ExecutionPolicy Bypass -File scripts/validation/check-all.ps1bash scripts/validation/check-all.shDoes 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.