agent-practices
AGENTS.md2 starsChanged 56 days agoMITView on GitHub
Maintain portable, high-signal guidance for humans and coding agents. This repository contains documentation and templates, not a product runtime. Run ./scripts/check.sh before proposing a change. It checks required files, skill metadata, template coverage, executable helpers, and trailing whitespace. If a check cannot run, say why in the pull request. Every contribution needs a clear user problem, copy-pasteable guidance, and a safe default. Claims about a tool's behavior should link to its official documentation or be phrased as tool-agnostic advice.
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 31,843 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 repository10 other files it gives its agents.
- code-reviewSkill
- coding-best-practicesSkill
- database-changesSkill
- debuggingSkill
- dependency-managementSkill
- documentationSkill
- git-workflowSkill
- release-managementSkill
- securitySkill
- testingSkill
The file
We split this AGENTS.md into the rules it gives an agent
6 rules. About 1 is common, 3 are distinctive, and 2 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 9
Read the relevant docs/, template, and skill before editing it.
- Distinctive: line 11
Prefer small, independently reviewable changes.
- Common: line 12
Never add secrets, private repositories, customer data, or instructions to defeat safeguards.
- Distinctive: line 13
Update examples when a convention changes.
- Very specific: line 17
Run ./scripts/check.sh before proposing a change.
- Distinctive: line 21
Every contribution needs a clear user problem, copy-pasteable guidance, and a safe default.
≈ “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
Repository operating contract
Purpose
Maintain portable, high-signal guidance for humans and coding agents. This repository contains documentation and templates, not a product runtime.
Working rules
- Read the relevant
docs/, template, and skill before editing it. - Preserve vendor neutrality. Describe capabilities, not endorsements.
- Prefer small, independently reviewable changes. Do not reformat unrelated files.
- Never add secrets, private repositories, customer data, or instructions to defeat safeguards.
- Update examples when a convention changes.
Validation
Run ./scripts/check.sh before proposing a change. It checks required files, skill metadata, template coverage, executable helpers, and trailing whitespace. If a check cannot run, say why in the pull request.
Review bar
Every contribution needs a clear user problem, copy-pasteable guidance, and a safe default. Claims about a tool's behavior should link to its official documentation or be phrased as tool-agnostic advice.
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
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.