skills/repo-hygiene-audit/SKILL.md
Inspect a GitHub repo for structural chaos, authority conflicts, and cleanup needs. Produce an audit report and remediation plan.
npx skillsauth add ariffazil/openclaw-workspace Repository Hygiene AuditInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Repos absorb debris over time: duplicate authority files, stale references, runtime artifacts, backup dumps, and constitutional leaks. This skill systematically inspects a repo and produces a cleanup plan.
| Input | Required | Description |
|-------|----------|-------------|
| repo_path | yes | Local path or owner/repo GitHub slug |
| repo_role | yes | Expected role: kernel, control-plane, execution, domain |
| depth | no | Audit depth: quick (top-level only), standard (3 levels), deep (full) |
Verify the repo does not claim authority it does not own:
CONSTITUTION.md, floors.py, judgment.py, vault999/) — must live in arifOS onlyREPO_ROUTING_CONSTITUTION.md) — must live in arifOS onlyarifos.init) — must live in arifOS onlyROOT_CANON.yaml) — must live in arifOS onlyIf found in non-arifOS repos → flag as constitutional leak.
Look for artifacts that do not belong in a source repo:
*backup*, *archive*, *old* at root).env with secrets, session JSON, memory dumps)node_modules/ at root (should not be committed)Verify directory structure matches the repo's canonical role:
| Role | Expected Structure |
|------|-------------------|
| kernel (arifOS) | core/, floors.py, judgment.py, vault999/, tests/ |
| control-plane (AAA) | src/, a2a/, agents/, contracts/, registries/, public/ |
| execution (A-FORGE) | src/, engine/, tools/, agents/, deploy/ |
| domain (GEOX/WEALTH/WELL) | src/, domain-specific engines, tests/ |
Flag phantom directories (referenced but empty) and missing canonical directories.
Scan for broken references caused by moves/renames:
import / require / from statements pointing to moved modulesIf repo is AAA or A-FORGE:
registries/skills.yaml must have a skills/<id>/SKILL.md| Tool | Purpose |
|------|---------|
| file-read | Read config and documentation |
| directory-list | Inspect structure |
| github-search | Search across repo for patterns |
| git-status | Check for uncommitted changes |
rm -rf or git push --force — these require Arif approval## Skill Result: repo-hygiene-audit
### Summary
[One-paragraph summary of repo health]
### Authority Boundary
- [ ] No constitutional leaks
- [ ] OR: Leaks found: [list]
### Runtime Debris
- [ ] No debris
- [ ] OR: Debris found: [list with sizes]
### Structural Issues
- [ ] Structure matches canonical role
- [ ] OR: Issues: [list]
### Broken References
- [ ] No broken refs
- [ ] OR: Broken refs: [list]
### Skill Registry
- [ ] Registry consistent
- [ ] OR: Orphan skills: [list]
### Recommendations
1. [Action] — [Priority] — [Owner]
2. [Action] — [Priority] — [Owner]
### Escalations
- None / [list]
| Condition | Escalate To | Method | |-----------|-------------|--------| | Constitutional leak found | arifOS 888_JUDGE | A2A verdict_request | | Secrets exposed | security.agent + Arif | Immediate HOLD + Telegram | | Cross-repo architectural change | Arif | 888_HOLD + human approval | | Deletion of >100 files proposed | Arif | 888_HOLD + human approval |
Skill version 1.0.0 — AAA Skill Library
development
Governed intelligence skill for AAA as the abstraction, attestation, and abduction control plane across arifOS, APEX, A-FORGE, GEOX, WEALTH, WELL, and the ariffazil profile repository. Use when the user asks to explain or design AAA, route agentic work, reduce chaos/entropy in an arifOS federation task, create AREP/task declarations, classify risk, plan multi-repo changes, review governance boundaries, or translate human intent into evidence-backed, authority-safe, recursively agentic workflows. Provides deterministic F1-F13 floor checking, bounded abduction, and FederationReceipt composition.
development
Check every skill’s “use when” and “do not use when” clauses for collisions, missing negatives, and vague verbs like “help,” “assist,” or “improve.” Load when linting, reviewing, or validating trigger boundaries.
development
Bootstrap, design, and package new skills. Load when capturing user intent for a new skill or drafting its initial instruction framework.
content-media
Diagnose which federation services are up, down, or drifting. Produce a prioritized remediation plan.