agentleFS
ariaxhanariaxhan/kernel-claude/skills/governance-sync/SKILL.md

governance-sync

Skill15 starsChanged 4 months agoMITView on GitHub

Audit or explicitly synchronize Claude and Codex repository governance without overwriting conflicts. Triggers: governance sync, CLAUDE.md, AGENTS.md, instruction mirror.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill 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:

    Use python3 "$KERNEL_GOVERNANCE_SYNC" init REPO --backup-dir REPO/.kernel-governance-backups/init only when the user explicitly wants …

    line 68

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
SKILL.md

Governance sync

This explicit-only operator audits native repository instructions and can create a missing Claude or Codex adapter from one declared source. It never edits a conflict.

Audit first

Resolve the installed script once. Claude Code and Codex compatibility loading provide CLAUDE_PLUGIN_ROOT; otherwise use the direct relative path from the documented installed skills/governance-sync working directory:

if [ -n "${CLAUDE_PLUGIN_ROOT:-}" ]; then
  KERNEL_GOVERNANCE_SYNC="$CLAUDE_PLUGIN_ROOT/scripts/governance-sync.py"
else
  KERNEL_GOVERNANCE_SYNC="../../scripts/governance-sync.py"
fi
test -f "$KERNEL_GOVERNANCE_SYNC"

Run python3 "$KERNEL_GOVERNANCE_SYNC" audit <root> --json. The audit discovers canonical Git roots, ignores caches, deduplicates linked worktrees, and reports: missing both, Claude-only, AGENTS-only, both identical, generated current, generated stale, incomplete generation, conflict, and scoped .claude states. Generated states are decided from manifest hashes and adapter provenance, not raw file equality. Missing-both is compliant and remains untouched unless init is explicit.

Writes require confirmation

Show the exact repository, source, target, and backup directory. Continue only after the user confirms one command:

python3 "$KERNEL_GOVERNANCE_SYNC" adopt REPO --source CLAUDE.md --backup-dir REPO/.kernel-governance-backups/adopt
python3 "$KERNEL_GOVERNANCE_SYNC" generate REPO --backup-dir REPO/.kernel-governance-backups/generate
python3 "$KERNEL_GOVERNANCE_SYNC" check REPO

AGENTS.md and .claude/CLAUDE.md are also valid sources. The manifest pins the source path, source hash, output, output hash, and generator version. Scoped sources stay where they are; only the missing root-native adapter is generated. A regular file conflict, unrecorded edit, stale hash, malformed manifest, symlink, or existing backup with different content stops the operation. Identical backups are idempotent. Backups stay inside the repository. Use separate BACKUPS/adopt and BACKUPS/generate phases so an approved source update preserves both the pre-adoption state and the adapter it replaces.

Writes are crash-consistent per file: each completed replacement is a whole, fsynced file. An interruption can leave some files current and others stale; check reports that drift without changing anything, and rerunning the write operation converges it. There is no background lock, journal, rollback, or cleanup of unknown temporary files. An existing backup-directory symlink is always rejected before path resolution.

Use python3 "$KERNEL_GOVERNANCE_SYNC" init REPO --backup-dir REPO/.kernel-governance-backups/init only when the user explicitly wants governance created in a repository where both native files are absent.

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.