agentleFS
Sign inStart free
SoonGwanSoonGwan/questionable-hires/AGENTS.md

questionable-hires

AGENTS.md14 starsChanged 55 days 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 31,439 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 repository9 other files it gives its agents.

The file

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

3 rules. About 0 are common, 0 are distinctive, and 3 are 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.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Treat benchmarks/featured.json as the single pointer to the landing-page benchmark.

    line 5
  2. Very specific:

    Keep README onboarding concise: link chronological per-tool experiments from the current status and dated reports instead of accumulating …

    line 8
  3. Very specific:

    Keep benchmarks/CURRENT-CANDIDATE-STATUS.md a concise dated decision index.

    line 9

≈ “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

Repository working agreements

  • Make small, focused commits as work progresses, as requested by the owner.
  • For banners, comparison charts, and other project visuals, use Montage semantic colors. Follow the palette and visual guidance in assets/README.md; do not introduce an unrelated palette unless the owner requests it.
  • Treat benchmarks/featured.json as the single pointer to the landing-page benchmark. Whenever featured benchmark data, charts, paths, or claims change, run python3 -B scripts/sync_featured_benchmark.py, review both README.md and README.ko.md, then run the same command with --check. Never update one language or a plotted number independently.
  • When a reviewed model result or material skill capability changes, refresh the dated checkpoint in benchmarks/CURRENT-CANDIDATE-STATUS.md and any affected English/Korean capability descriptions. Link the measured resource and limitations; do not relabel old measurements or promote a single favorable pair into the featured benchmark. Keep historical checkpoints explicitly historical, and separate local tests from model evidence and hosted release checks.
  • Label non-featured checkpoint summaries with their checkpoint ID, date and measured resource rather than an independently maintained “latest” claim. Keep adverse results and limitations accessible when reorganizing onboarding; synchronize both README languages without changing frozen charts.
  • Keep README onboarding concise: link chronological per-tool experiments from the current status and dated reports instead of accumulating each development step in the landing page. Preserve the dated docs/DEVELOPMENT-NOTES*.md snapshots; keep mixed/adverse evidence and limitations visible in both README languages.
  • Keep benchmarks/CURRENT-CANDIDATE-STATUS.md a concise dated decision index. Preserve superseded checkpoints in same-directory dated history snapshots so relative evidence links remain valid. Before repeating an experiment or adding analysis tooling, search histories and existing scripts for the same approach; reused development cases are not independent validation.

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