i-dont-believe-you
AGENTS.md18 starsChanged 54 days agoMITView on GitHub
The product is skills/verify-before-claiming/SKILL.md. Everything else exists to prove it works and to get it installed. This repository ships a skill that forbids unverified success claims. Apply it to your own work in it. Run the eight commands in SKILL.md against your diff before you report that anything passes, and paste the output. Every bash block that is a check needs a <!-- check: id --> marker, or the evals will not see it and tests/skill.test.ts will fail. Run npm…
- Installs packages
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 33,315 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 repository4 other files it gives its agents.
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 43
Every bash block that is a check needs a <!-- check: id --> marker, or the evals will not see it and tests/skill.test.ts will fail.
≈ “Common” is an estimate: at least 1 in 5 of the 42,283 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
Working in this repository
The product is skills/verify-before-claiming/SKILL.md. Everything else exists to prove it works
and to get it installed.
Before you claim anything here
This repository ships a skill that forbids unverified success claims. Apply it to your own work
in it. Run the eight commands in SKILL.md against your diff before you report that anything
passes, and paste the output.
Commands
npm install
npm run evals # runs the SKILL.md commands against throwaway git repos
npm test # harness and optional CLI
npm run lint
npm run typecheck
npm run sync-skill # after editing SKILL.md
npm run sync-skill:check # what CI runsLayout
| Path | What it is |
skills/verify-before-claiming/SKILL.md | the product |
evals/cases/<name>/ | before/, after/, claim.txt, expected.json |
evals/skill.ts | extracts the commands out of SKILL.md |
src/ | optional CLI running the same checks as a CI gate |
.claude/ .codex/ .cursor/ .opencode/ .agents/ | generated copies, do not edit by hand |
Rules for a check
- One shell command, silent when there is no problem.
- Nothing beyond
git,grep,sed,command bash. - Under a second on a large repository.
- A false positive is worse than a miss.
If you edit SKILL.md
Every bash block that is a check needs a <!-- check: id --> marker, or the evals will not see
it and tests/skill.test.ts will fail. Run npm run sync-skill afterwards.
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.

