sebastienrousseausebastienrousseau/agtmls/AGENTS.md

agtmls

AGENTS.md1 starsChanged 3 months agoApache-2.0View on GitHub

Invariants for AI-assisted contributions to AgtMLS. Read this before changing anything. All contributions to AgtMLS MUST comply with the global standards located at /Users/seb/Code/: - /Users/seb/Code/AGENTS.md (Code Repositories Standard) - /Users/seb/Code/GEMINI.md (Global Standard & Conventions) - /Users/seb/Code/PR-TEMPLATE.md (Pull Request Specification) - /Users/seb/Code/README-TEMPLATE.md (Canonical README Standard) - /Users/seb/Code/REPO-STANDARD.md (Repository Gold Standard) The gate is every check in checks.json, plus the unit tests. Before concluding any task, run: All unit tests, behavioral evals, routing checks, and doctor inspections must pass with 0…

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 repository19 other files it gives its agents.
- agent-loop-designSkill
- ai-supply-chain-securitySkill
- anti-slop-pr-and-writingSkill
- authoring-portable-skillsSkill
- brainstormingSkill
- cross-language-portSkill
- giving-code-reviewSkill
- handoffSkill
- incident-responseSkill
- pr-review-and-releaseSkill
- receiving-code-reviewSkill
- refactoring-safelySkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- using-agtmlsSkill
- verification-before-completionSkill
- vetting-a-skill-before-installSkill
- web-research-and-source-triageSkill
- writing-plansSkill

The file

We split this AGENTS.md into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
All contributions to AgtMLS MUST comply with the global standards located at /Users/seb/Code/:
line 10
Very specific:
Zero external dependencies: All scripts in scripts/ must remain stdlib-only so that uvx agtmls and standalone script invocations are …
line 20
Distinctive:
All unit tests, behavioral evals, routing checks, and doctor inspections must pass with 0 failures and 0 warnings.
line 32
Very specific:
Before any task, run the hygiene check of ~/Code/AGENTS.md section 0 (open Dependabot PRs and alerts, code-scanning and secret-scanning …
line 43

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

AGENTS.md

AGENTS.md

Invariants for AI-assisted contributions to AgtMLS. Read this before changing anything.

Master Standards & Authoritative Sources of Truth

All contributions to AgtMLS MUST comply with the global standards located at /Users/seb/Code/:

/Users/seb/Code/AGENTS.md (Code Repositories Standard)
/Users/seb/Code/GEMINI.md (Global Standard & Conventions)
/Users/seb/Code/PR-TEMPLATE.md (Pull Request Specification)
/Users/seb/Code/README-TEMPLATE.md (Canonical README Standard)
/Users/seb/Code/REPO-STANDARD.md (Repository Gold Standard)

The core invariants

Strict SemVer sequencing policy: Public releases stay on the 0.0.x line and increment strictly by 0.0.1. Never manually edit version numbers; use python3 scripts/bump-version.py. v0.1.0 is forbidden until v0.0.999 exists.
Zero external dependencies: All scripts in scripts/ must remain stdlib-only so that uvx agtmls and standalone script invocations are instant without resolving packages.
Never hand-edit generated artifacts: Artifacts including index.json, CATALOG.md, site/index.html, mcp-resources.json, SBOM.spdx.json, provenance.json, and plugin manifests are derived. Regenerate them with their respective generator scripts (--write).
Preserve dual licensing: The repository is dual-licensed under Apache-2.0 OR MIT. All files must declare an SPDX license header.

Before you claim to be done

The gate is every check in checks.json, plus the unit tests. Before concluding any task, run:

python3 scripts/run-all-checks.py

All unit tests, behavioral evals, routing checks, and doctor inspections must pass with 0 failures and 0 warnings. State what you ran and what you observed.

Unit-test coverage of scripts/ and src/ (lines and branches, measured from the unit suite alone) MUST stay at or above 98%, and the library core at 100%:

python3 scripts/run-coverage.py --scope unit
python3 scripts/run-coverage.py --scope core

Both need coverage installed; CI runs them in conformance.yml. New code ships with tests that were seen to fail before the change.

Before any task, run the hygiene check of ~/Code/AGENTS.md section 0 (open Dependabot PRs and alerts, code-scanning and secret-scanning alerts, red CI on main), and fix what it finds first. Every function in scripts/, src/ and fuzz/ is held to cyclomatic 10, cognitive 15, Halstead difficulty 30 and 60 lines, and every file to 500 lines, by scripts/check-complexity.py in the gate. Code you write or touch meets them; the offenders recorded in complexity-baseline.json may only shrink.

Download now (ZIP)

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

What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free