agentleFS
hurttlockerhurttlocker/o8/AGENTS.md

o8

AGENTS.md24 starsChanged 42 days agoMITView on GitHub

These are the canonical operating rules for agents working in this repository. Read the sections that apply to the current task once. Load scoped references only when their trigger applies. - Turn the request into an observable outcome. Separate symptoms, established facts, hypotheses, and root cause. Choose the smallest complete remedy inside the requested scope. - Read-only or diagnostic work stays non-mutating. A change request includes implementation, proportionate verification, a focused commit, push, and pull request. An operator or lead…

  • 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 repository11 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, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read-only or diagnostic work stays non-mutating.

    line 10
  2. Distinctive:

    If blocked, preserve state and report the exact blocker, evidence, and shortest safe unblock.

    line 20
  3. Distinctive:

    Read only matching skills before the relevant work:

    line 25

≈ “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

o8 Repository Rules

These are the canonical operating rules for agents working in this repository. Read the sections that apply to the current task once. Load scoped references only when their trigger applies.

Outcome ownership and authority

  • Turn the request into an observable outcome. Separate symptoms, established facts, hypotheses, and root cause. Choose the smallest complete remedy inside the requested scope.
  • Read-only or diagnostic work stays non-mutating. A change request includes implementation, proportionate verification, a focused commit, push, and pull request. An operator or lead agent may merge a reviewed CI-green pull request under standing authorization; implementation workers and packets never self-approve or merge. Version bumps, releases, production changes, public posts, destructive actions, spending, credentials, and another person's machine remain operator gates unless the current task explicitly authorizes them.
  • Preserve dirty worktrees and unrelated edits. Inspect status before editing, use explicit pathspecs, and never discard, stash, reset, or rewrite work you do not own.
  • A plan, test, commit, merge, or delivered message is evidence, not automatic proof. Verify through the real entry point and report Outcome, Evidence, Residual, and Decision.
  • If blocked, preserve state and report the exact blocker, evidence, and shortest safe unblock.

Verification and skills

For code changes, run npm run typecheck:fast and the required completion gates; use focused checks while iterating. Read only matching skills before the relevant work:

  • Code, UI/TSX, checks.
  • Issues/PRs/commits.
  • Agents/high-risk review/CLI, release.

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.