agentleFS
Sign inStart free
Nisus74Nisus74/humanise/AGENTS.md

humanise

AGENTS.md1 starsChanged 5 months agoMITView on GitHub

This repo is a writing skill, not an app. If you are an AI coding agent (Codex and others read this file), apply the humanise skill whenever you write or edit prose for the user: commit-message bodies aside, anything human-facing (READMEs, docs, posts, emails, PR descriptions). How to use it: The rules in skill/references/ are the shared engine; the voice in skill/profile/ is the user's. Don't apply the prose sweep to code, config, or quoted material (see the "When to…

  • 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 31,700 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 repository3 other files it gives its agents.

The file

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

8 rules. About 0 are common, 3 are distinctive, and 5 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.

  • 5

    Very specific

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

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

    Read skill/SKILL.md for the workflow (assess, draft from the user's corpus, two-pass sweep, self-critique, verify).

    line 7
  2. Very specific:

    Read the user's skill/profile/ (their soul.md, voice-fingerprint.md, absolute-rules.md) if present.

    line 8
  3. Very specific:

    Run the checker on anything substantial: python3 skill/evals/assertions/writing_checks.py <draft-file> <dialect> [medium].

    line 9
  4. Distinctive:

    Never invent testimonials, metrics or compatibility claims.

    line 17
  5. Very specific:

    Keep privacy language technically precise.

    line 18
  6. Distinctive:

    Do not change the engine during a launch-surface task.

    line 20
  7. Very specific:

    Run npm run quality before completion.

    line 21
  8. Distinctive:

    Do not add dependencies without explicit approval.

    line 22

≈ “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

AGENTS.md

This repo is a writing skill, not an app. If you are an AI coding agent (Codex and others read this file), apply the humanise skill whenever you write or edit prose for the user: commit-message bodies aside, anything human-facing (READMEs, docs, posts, emails, PR descriptions).

How to use it:

  1. Read skill/SKILL.md for the workflow (assess, draft from the user's corpus, two-pass sweep, self-critique, verify).
  2. Read the user's skill/profile/ (their soul.md, voice-fingerprint.md, absolute-rules.md) if present. If there's no profile/, tell the user to follow docs/getting-started.md and fall back to the engine defaults.
  3. Run the checker on anything substantial: python3 skill/evals/assertions/writing_checks.py <draft-file> <dialect> [medium].

The rules in skill/references/ are the shared engine; the voice in skill/profile/ is the user's. Don't apply the prose sweep to code, config, or quoted material (see the "When to apply, and when to hold off" section in skill/SKILL.md).

Engine changes clear the acceptance gate in skill/evals/self-harness-loop.md, runnable end to end via /humanise improve. Never tune a change against skill/evals/holdout-evals.json (reserved held-out), and treat skill/profile/learning/ as soul: the user's verbatim edit history, never committed.

Launch documentation

  • Never invent testimonials, metrics or compatibility claims.
  • Keep privacy language technically precise. Separate what Humanise stores locally from what the selected AI host may receive under its own data policy.
  • Do not change the engine during a launch-surface task.
  • Run npm run quality before completion.
  • Do not add dependencies without explicit approval.

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