cd /news/ai-agents/i-labeled-558-agents-md-files-here-s… · home topics ai-agents article
[ARTICLE · art-129023] src=dev.to ↗ pub= topic=ai-agents verified=true sentiment=· neutral

I labeled 558 AGENTS.md files. Here's what they say — and what almost nobody writes down

A developer collected 558 public AGENTS.md files and labeled them against a nine-category taxonomy using an auditable rule-based classifier, achieving 92% precision and 70% recall on 55 held-out English files and 88%/73% on 50 Chinese files. The most common categories were prohibitions (85.7%) and build/test commands (82.8%), while gotchas (13.6%) and rules about agent behavior (25.8%) were rarest. The labeled dataset and an accompanying agent-charters tool were released so the numbers can be recomputed.

by read4 min views1 publishedSep 14, 2026

TL;DR — I collected 558 AGENTS.md files from public repos and labeled each one against a 9-category

taxonomy with a rule-based classifier (no LLM in the loop, so it is auditable and recomputable). Then I

blind-labeled held-out samples and compared: 92% precision / 70% recall on 55 English files,

88% / 73% on 50 Chinese files. The most common categories are prohibitions (85.7%) and build/test

commands (82.8%). The rarest: gotchas (13.6%) and instructions about how the agent itself should behave

(25.8%).

Almost every discussion about AGENTS.md is anecdote-led: my repo's file works, my agent ignores it,

a good one is a model upgrade, a bad one is worse than nothing. All of that may be true — but nobody

seems to have the distribution. So I built it: snapshot of 558 files from 558 public repos

(2026-09-10, 5.3 MB, 516 usable for statistics), labeled, versioned, and published with the tooling.

Nine categories: boundaries, build_test, workflow, structure, style, environment, overview,

agent_meta (rules about the AI itself), gotchas. Labeling is done by pattern rules over headings and

body text — deliberately, because a rule set can be read, argued with, and re-run, and every number below

can be recomputed from the released dataset. I then measured how well the rules match a human reading:

100 files in-sample (upper bound, 90%/75%) and two held-out sets I had never tuned against —

55 English (92%/70%) and 50 Chinese (88%/73%). Held-out numbers use the conservative reading

(items I was unsure about count as classifier errors).

category share of 516 files
boundaries (what you must never do) 85.7%
build_test (install/build/test/CI commands) 82.8%
workflow (branching, commits, review, release) 67.1%
structure 59.1%
style 54.5%
environment 45.0%
overview 32.2%
agent_meta 25.8%
gotchas 13.6%

The 2.9 pp gap between the top two is smaller than the known false-positive rate (~3%) of the

prohibition pattern — so the honest statement is tied for first, not "prohibitions beat build commands".

gotchas is dead last at 13.6%. Worse: when people do open a "known issues" section, a third of it

isn't a gotcha. I hand-read 120 items from those sections:

That 8% is the part an agent can never derive from the code — and it is exactly the part that is

almost never written down.

workflow In a controlled experiment (11 repos × 3 prompt styles), prompts that listed topics explicitly produced

9/9 categories, while prompts that left the slots implicit skipped workflow in 11 out of 11 files.

Point at workflow by name and it appears 3/3 times, with real content. The gap is not knowledge,

it is questions — which is why I turned the corpus distribution into a checklist tool.

English files fail differently from Chinese ones. English: gotchas recall 32–38% — the classifier

misses casual "watch out" prose. Chinese: agent_meta recall 26% — Chinese files express agent rules

in the second person ("you are the dispatcher, not the executor"), and the body-pattern rules for that

category are entirely English, so the whole style is invisible to them. File-level exact agreement

(9/9 categories identical) is 12% in both languages.

49% point to some other file; 15% route to a knowledge or rules directory. That's a structural

fact about the format, and it means "does this repo have an AGENTS.md?" is a much weaker question than

"what is actually in it".

pip install agent-charters

agent-charters brief      # checklist of the 9 slots + a paste-ready prompt
agent-charters compare your-AGENTS.md   # your coverage vs the 558-file baseline
agent-charters refs your-AGENTS.md      # does your file point at paths that exist

Honest note: compare is a checklist, not an oracle. It warned me that one of the nine categories was

missing from a file I wrote myself — it was actually present, but the heading used the tool's own slot name

instead of natural language. That is documented in the repo (along with the exact experiment) rather than

quietly patched, because a tool that tells you "you're missing X" should be checked by a human.

LIMITATIONS.md in the repo I'm looking for 2–3 people who are not me to run compare on an AGENTS.md they actually maintain and

tell me where it's wrong — missing a category you clearly have, or claiming one you don't. That is the one

piece of evidence this project doesn't have yet: an external user. Issues and comments are both fine.

── more in #ai-agents 4 stories · sorted by recency
── more on @agent-charters 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/i-labeled-558-agents…] indexed:0 read:4min 2026-09-14 ·