agentsclimarketplace

Cleanup

Skill ronniepinnell/casper/collection/verification-and-audit/cleanup

👻 The friendly ghost in your git. Your AI said done — Casper makes it prove it. Claim-evidence hooks + a verdict ledger for Claude Code.

Install
npx -y skills add ronniepinnell/casper --skill cleanup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Mass reconciliation when things are out of whack. Audit, triage, fix, verify.

SKILL.md

1.7 KB, 362 tokens by cl100k_base, as published. Nobody here has run it

/cleanup — Mass Reconciliation

For when docs, specs, phase tables, or code are out of sync. Audit everything, triage, fix.

What Happens

  1. Audit: Run parallel agents to check:
    • Doc staleness (config/doc_mappings.json)
    • Phase table completeness (captured ideas via storage list_ideas → plan rows)
    • Spec coverage (features in specs → phase table rows)
    • Code-spec drift (decision-store entries via storage list_decisions → actual code)
    • Dead references (links to files that don't exist)
  2. Triage: Present findings to Operator with severity (P0/P1/P2)
  3. Generate fix prompts: Through Gate 2 for validation
  4. Execute: Fix issues (with Operator approval for P0s)
  5. Verify: Re-run audit to confirm fixes

Team Mode (if CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS is set)

Audit team: {agents.completion_audit} + {agents.spec_audit} + compliance-checker + {agents.doc_sync} agents in parallel.

Fallback Mode (if env var not set)

Sequential subagent calls: {agents.completion_audit} → {agents.spec_audit} → compliance-checker → {agents.doc_sync}.

Gate References

  • All fix prompts go through Gate 2 (pre-commit)
  • All pushes go through Gate 1 (pre-push content pipeline)

Judgment weave (see /judgment)

  • Audit phase: big messes are /sweep jobs — fan out rather than spot-check.
  • "All clean" claim: run /refute before declaring the reconciliation done.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.