agentleFS
Sign inStart free
emlaiemlai/defi-native-skill/AGENTS.md

defi-native-skill

AGENTS.md44 starsChanged 2 months agoMITView on GitHub

defi-native is a public Agent Skill (SKILL.md + references/) plus its marketing site (site/, deployed to https://defi-native.ai by GitHub Pages on every push to main). This file is the operating manual for any coding agent. CONTRIBUTING.md covers content rules for the skill body; MAINTENANCE.md covers the upkeep policy. - Never use em or en dashes. Commas, colons, periods, parentheses. - Expand every acronym at first use per file. - Every number carries an as-of date or is labeled a calibration…

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 2

    1. BankrBot/openclaw-skills★ 1.2k100% same
    2. BankrBot/skills★ 1.2k100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • 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 31,774 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 repository2 other files it gives its agents.

The file

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

5 rules. About 0 are common, 3 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never use em or en dashes.

    line 11
  2. Distinctive:

    Every number carries an as-of date or is labeled a calibration example.

    line 13
  3. Distinctive:

    Never invent a metric, count, or quote.

    line 14
  4. Very specific:

    After ANY change to site/assets/site.css or site.js run python3 scripts/build_site.py --stamp to refresh the cache-busting hashes on both …

    line 43
  5. Very specific:

    Update: SKILL.md version, manifest.json version and counts, api-routes.json version, CHANGELOG.md, README.md counts, llms.txt counts, site …

    line 61

≈ “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: how to work on this repo (any agent, any harness)

defi-native is a public Agent Skill (SKILL.md + references/) plus its marketing site (site/, deployed to https://defi-native.ai by GitHub Pages on every push to main). This file is the operating manual for any coding agent. CONTRIBUTING.md covers content rules for the skill body; MAINTENANCE.md covers the upkeep policy.

Hard conventions, every edit, no exceptions

  • Never use em or en dashes. Commas, colons, periods, parentheses.
  • Expand every acronym at first use per file.
  • Every number carries an as-of date or is labeled a calibration example.
  • Never invent a metric, count, or quote. Verify before writing.
  • SKILL.md stays lean; knowledge lives in references/ and loads per task.
  • The skill is read-only by design: it never constructs, signs, submits, or approves transactions. Do not weaken that language anywhere.

Versioning (MAINTENANCE.md is canonical)

Patch = calibration refresh. Minor = new section or reference file. Major = structural change to directives or workflow. Every release gets a CHANGELOG.md entry, and the site changelog gets a plain-language entry.

Gates: run these before any push

python3 scripts/verify_manifest.py --skip-network   # skill invariants, counts, caps
python3 scripts/build_site.py                       # site must match the skill

Both must exit 0. CI runs them on every PR (.github/workflows/pr-check.yml) and the Pages deploy refuses drifted pages. What build_site.py enforces: stat tiles match manifest/api-routes/glossary/evals counts, version strings match SKILL.md on both pages, the update page offers the current version, every must or core priority manifest source is named in the data-sources list, and asset URLs carry current content hashes.

Editing the site

  • Pages: site/index.html and site/update/index.html. Shared styles and script: site/assets/site.css and site/assets/site.js.
  • After ANY change to site/assets/site.css or site.js run python3 scripts/build_site.py --stamp to refresh the cache-busting hashes on both pages, then the gate. Skipping this fails the deploy.
  • Source logos for the marquee live in site/assets/logos/.
  • The GitHub star count is fetched live client side; the number in the HTML is only a no-JS fallback.

Workflow

Branch, commit, push, open a PR, wait for the verify check, merge. Merging to main deploys the site automatically. Do not push to main directly. Do not commit anything from ../ (the parent folder holds private research and memory; .gitignore guards it, leave those guards alone). Show Emily before anything user-facing ships if she has not already specified the exact copy.

Skill releases

Update: SKILL.md version, manifest.json version and counts, api-routes.json version, CHANGELOG.md, README.md counts, llms.txt counts, site stat tiles and site changelog entry, the update page demo line. The gates catch most of these; the CHANGELOG entry and site changelog prose are on you.

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