agentleFS
Sign inStart free
darshitppdarshitpp/x-agent/AGENTS.md

x-agent

AGENTS.md2 starsChanged 6 months agoMITView on GitHub

x-agent is a set of agentskills.io skills, one per supported CLI. Each <cli>/SKILL.md is a thin entry point that points at shared references. No application code; the product is Markdown plus a maintainer shell wrapper. This file lists no CLIs or files that change. Look them up: - Supported CLIs: ls */SKILL.md (each top-level dir with a SKILL.md is one CLI). - Everything that must change for a CLI: git grep -n -i <existing-cli-name> (use a similar, recent CLI such…

  • 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 31,843 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 repository10 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, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    No self-call detection: invoking the same CLI with a different model is a supported cross-validation use case.

    line 25
  2. Very specific:

    Document only flags you verified: Check the installed CLI is current (npm view <pkg> version, or npx -y <pkg>@latest --help.

    line 26
  3. Very specific:

    Read the test setup in README.md and .github/workflows/run-tests.yml for the current commands and dependencies (pytest and bats.

    line 43

≈ “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

x-agent is a set of agentskills.io skills, one per supported CLI. Each <cli>/SKILL.md is a thin entry point that points at shared references. No application code; the product is Markdown plus a maintainer shell wrapper.

Discover, don't assume

This file lists no CLIs or files that change. Look them up:

  • Supported CLIs: ls */SKILL.md (each top-level dir with a SKILL.md is one CLI).
  • Everything that must change for a CLI: git grep -n -i <existing-cli-name> (use a similar, recent CLI such as the latest git log --stat addition as the template) and mirror each hit.
  • Layout and CI details: README.md (directory tree, CI section) and .github/workflows/.

Rules

  • references/ and assets/ at the repo root are canonical. Each skill folder holds byte-identical copies, because skills install standalone. After editing a canonical file, copy it into every skill folder that has it. pytest fails on drift and prints the cp command.
  • SKILL.md stays thin: frontmatter plus "read shared-procedure, read the CLI reference, follow it". Flags, models and version info belong in references/cli-<cli>.md.
  • Frontmatter limits are enforced by scripts/validate-metadata.py (run it with --help).
  • scripts/query-cli.sh is maintainer-only and not shipped; skills never call it.
  • No self-call detection: invoking the same CLI with a different model is a supported cross-validation use case.
  • Document only flags you verified. Check the installed CLI is current (npm view <pkg> version, or npx -y <pkg>@latest --help; for others use the CLI's own update command), then read <cli> --help; otherwise cite the public docs. Mark anything unverified as such in the reference.
  • Changes go through a PR; CI runs integrity, pytest and bats on Ubuntu and macOS.

Adding a CLI or changing its flags

A CLI is touched in five places; find the exact current files with git grep as above rather than trusting a list:

  1. Canonical reference references/cli-<cli>.md, plus its copy in <cli>/references/.
  2. Skill folder <cli>/ (SKILL.md and copies of the shared procedure and result template).
  3. The matching case in scripts/query-cli.sh.
  4. bats tests for that case, and the CLI-name lists in the pytest files.
  5. README mentions of the CLI.

Testing

Read the test setup in README.md and .github/workflows/run-tests.yml for the current commands and dependencies (pytest and bats; scripts/run-workflow.sh runs CI locally via act).

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