agentleFS
Sign inStart free
Gord1yGord1y/countersign-skills/CLAUDE.md

countersign-skills

CLAUDE.md2 starsChanged 38 days agoMITView on GitHub

Skills, global rule modules and subagent types for coding agents (Claude Code, Codex, Antigravity), installed by POSIX sh scripts (install.sh, update.sh, uninstall.sh, with lib.sh shared) and described by a generated catalog.json. jq handles the JSON. - Zero comments in shell scripts, workflows and Markdown. Names carry the meaning. Rationale goes into docs/. A pinned action's # vX.Y.Z suffix stays. - The only # shellcheck directive is source=lib.sh before each . "$root/lib.sh". Fix the code otherwise, or use the exclusions in…

The file

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

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Every SKILL.md ends with the exact local.md line, keeps description at most 200 bytes and its folder name equal to its name, and has …

    line 14

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
CLAUDE.md

countersign-skills

Skills, global rule modules and subagent types for coding agents (Claude Code, Codex, Antigravity), installed by POSIX sh scripts (install.sh, update.sh, uninstall.sh, with lib.sh shared) and described by a generated catalog.json. jq handles the JSON.

Rules

  • Zero comments in shell scripts, workflows and Markdown. Names carry the meaning. Rationale goes into docs/. A pinned action's # vX.Y.Z suffix stays.
  • The only # shellcheck directive is source=lib.sh before each . "$root/lib.sh". Fix the code otherwise, or use the exclusions in .shellcheckrc and scripts/lint.sh.
  • POSIX sh with set -eu, BSD and GNU compatible, macOS first.
  • Every SKILL.md ends with the exact local.md line, keeps description at most 200 bytes and its folder name equal to its name, and has when_to_use unless it disables model invocation.
  • List every skill in skills.tsv and every rule in rules.tsv. Bump a skill's version there with every change to it.
  • Regenerate catalog.json with scripts/catalog.sh; ./test.sh fails while it is stale.
  • Prose: short declarative sentences, English only.
  • Conventional Commits, one task per commit.

Gates

scripts/check.sh

It must pass before a change is done. It runs ./test.sh, every scripts/test-*.sh and scripts/lint.sh.

Repo facts

FactDetail
The surfaces, and how to reach eachcli: run install.sh, update.sh and uninstall.sh against a scratch home, HOME="$(mktemp -d)" CLAUDE_CONFIG_DIR= ./install.sh, never the real one. bin/statusline: reads Claude Code's status line JSON on stdin. bin/gh: wraps gh; run it with a stub gh first on PATH, as scripts/test-gh.sh does.
BranchesOne branch per release carries all of its work and its note, and goes into main through one pull request. There is no staging.
ReviewsNo automated review and no required approval; reviews follow docs/review-checklist.md.
Release PRsThe release branch goes into main as a merge commit titled chore(release): v<semver>, with no promotion contract; the note is releases/release-<semver>.md (contract in releases/README.md) and is committed on that branch, and a v* tag on main starts the release workflow. No check runs only on release PRs.
LicenceMIT, no CLA.

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