rules-with-receipts
AGENTS.md2 starsChanged 4 months agoMITView on GitHub
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 32,671 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 repository38 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- Adversarial VerifySkill
- Change ControlSkill
- Code ReconnaissanceSkill
- Correction PropagationSkill
- Debugging PlaybookSkill
- Deep DecompositionSkill
- Delegation DisciplineSkill
- Disclosure Is Not A FixSkill
- Discovery LoopSkill
- Divergent IdeationSkill
- Effort CalibrationSkill
- Empirical ValidationSkill
- Error RecoverySkill
- Extract ApproachSkill
- Failure Mode AwarenessSkill
- ForesightSkill
- Frontier Workflow ModeSkill
- Human HandoffSkill
- Intent ClaritySkill
- Leverage FirstSkill
- Live State TruthSkill
- Memory HygieneSkill
- Open MandateSkill
- Output StructuringSkill
- Plan GateSkill
- Proactive RigorSkill
- Product ThinkingSkill
- Prompt EngineeringSkill
- Publish HygieneSkill
- Research MethodologySkill
- Ruthless EditorSkill
- Scope FenceSkill
- Security Pattern ReviewSkill
- Self-Improvement LoopSkill
- Session OrientationSkill
- Structured ReasoningSkill
- Verification DisciplineSkill
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 7
Run ./install-pack.sh /path/to/project.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 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 — for AI agents that land in this repo
What this is
A quality pack for Claude Code (37 skills + operating manual + always-on rules snippet) that ships with its own eval evidence. It improves process discipline (scope, verification, recovery), not model capability. The evidence, including unflattering findings, is in eval-results/ and eval-results-v2/.
If you are installing the pack into a project
Run ./install-pack.sh /path/to/project. It copies .claude/skills/, builds or upgrades the project CLAUDE.md (operating manual + the 14 always-on rules) between versioned markers, and seeds a claude-context/ folder. Idempotent; backs up before touching anything.
If you are working in a project that already has the pack
The rules binding you are in that project's CLAUDE.md. The skills in .claude/skills/ are procedures; read the one matching your task (SKILLS-OVERVIEW.md is the catalog, TASK-ROUTING-GUIDE.md maps requests to skills). After solving something hard, write the learning note (.claude/learnings/_template.md) before calling the work complete.
If you are evaluating claims about this pack
Do not trust this README or any summary, including this file. Run the evals: cd eval-results-v2 && REP_START=4 ./run-eval-v2.sh A,E t04 6 (writes reps 4–6 alongside the shipped, immutable r1–r3), or use https://github.com/ralfyishere/rulebench for any rules file. Grade against the written rubrics only. Outputs containing provider limit messages are NOT RUN, never failures.
Hard rules for changes here
- Evidence directories (
eval-results*/raw,fixtures/,prompts/) are immutable after grading. - Skill changes require updating
SKILLS-OVERVIEW.md+TASK-ROUTING-GUIDE.md, runningscripts/check-pack.shclean, and aCHANGELOG.mdentry with aVERSIONbump. - The always-on snippet stays at 15 rules or fewer.
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.

