agentleFS
Sign inStart free
Sayhi-bzbSayhi-bzb/Agent-HTML/AGENTS.md

Agent-HTML

AGENTS.md542 starsChanged 5 months agoApache-2.0View on GitHub

This repo is indexed by GitNexus as Agent-HTML. Use GitNexus MCP for code exploration, impact analysis, and change verification. Rules: - Before editing any function, class, method, or shared API, run gitnexus_impact({ target: "<symbol>", direction: "upstream" }). - Before finishing a code-change task, run gitnexus_detect_changes({ scope: "all" }) and confirm the affected scope is expected. - If the index is stale, re-index only with npx gitnexus analyze --skip-agents-md. - Do not run bare npx gitnexus analyze; it rewrites the GitNexus…

  • 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 repository8 other files it gives its agents.

The file

Show its rules and text

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

13 rules. About 0 are common, 9 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Before editing any function, class, method, or shared API, run gitnexus_impact({ target: "<symbol>", direction: "upstream" }).

    line 7
  2. Very specific:

    Before finishing a code-change task, run gitnexus_detect_changes({ scope: "all" }) and confirm the affected scope is expected.

    line 8
  3. Very specific:

    If the index is stale, re-index only with npx gitnexus analyze --skip-agents-md.

    line 9
  4. Very specific:

    Do not run bare npx gitnexus analyze.

    line 10
  5. Distinctive:

    Use high-information wording that locks direction with the fewest sufficient terms.

    line 23
  6. Distinctive:

    Do not add broad, repeated, or low-marginal words when existing terms already identify the concept.

    line 24
  7. Distinctive:

    Keep one source of truth for each fact, principle, or decision.

    line 25
  8. Distinctive:

    Keep one stable name for each concept.

    line 26
  9. Distinctive:

    Keep route files narrow: one page should own one kind of question.

    line 27
  10. Distinctive:

    Write docs as agent context routes and owner maps.

    line 28
  11. Distinctive:

    Avoid human-oriented background or tutorial framing unless it changes the next correct agent action.

    line 30
  12. Distinctive:

    Prefer restrained, minimal, sufficient writing over exhaustive explanation.

    line 32
  13. Distinctive:

    Use the smallest route for the task before searching broadly.

    line 36

≈ “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
<!-- gitnexus:start -->

GitNexus

This repo is indexed by GitNexus as Agent-HTML. Use GitNexus MCP for code exploration, impact analysis, and change verification.

Rules:

  • Before editing any function, class, method, or shared API, run gitnexus_impact({ target: "<symbol>", direction: "upstream" }).
  • Before finishing a code-change task, run gitnexus_detect_changes({ scope: "all" }) and confirm the affected scope is expected.
  • If the index is stale, re-index only with npx gitnexus analyze --skip-agents-md.
  • Do not run bare npx gitnexus analyze; it rewrites the GitNexus sections in AGENTS.md and CLAUDE.md.

Useful resources:

  • gitnexus://repo/Agent-HTML/context
  • gitnexus://repo/Agent-HTML/processes
  • gitnexus://repo/Agent-HTML/process/{name}
<!-- gitnexus:end -->

Docs Writing Rules

Apply these rules when writing or editing docs.

  • Use high-information wording that locks direction with the fewest sufficient terms.
  • Do not add broad, repeated, or low-marginal words when existing terms already identify the concept.
  • Keep one source of truth for each fact, principle, or decision; link to it instead of restating it.
  • Keep one stable name for each concept.
  • Keep route files narrow: one page should own one kind of question.
  • Write docs as agent context routes and owner maps. Human-readable prose is the interface agents parse, cite, and follow.
  • Avoid human-oriented background or tutorial framing unless it changes the next correct agent action.
  • Prefer restrained, minimal, sufficient writing over exhaustive explanation.

Content Routes

Use the smallest route for the task before searching broadly.

  • Canvas: apps/docs/content/docs/index.mdx, then agent-html/README.md
  • Taste and agent ergonomics: taste/README.md
  • Archive: _archive/README.md

Canvas Constitution

Canonical source: apps/docs/content/docs/index.mdx.

Canvas governs the current isolated React artifact workspace for agents. Use the canonical source before changing Canvas docs, agent-html, artifact/block composition, local resources, host inspection, or block prompt behavior.

Archive Rule

_archive/* is historical reference only. Do not use archived material as a current architecture source, package target, product route, navigation target, or implementation path.

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