agentsclimarketplace

Audit collection

Skill gnomatix/steering/.claude/skills/audit-collection

First-pass version of a cross-vendor "steering file" auditor / linter "suite"; immediately runnable via provided scripts using 'claude' CLI. #yourewelcome

Install
npx -y skills add gnomatix/steering --skill audit-collection

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

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 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

Audit a body of related agent documents — a `.kiro/` tree, a `~/.claude/skills/` directory, a project's `docs/agent/` collection, or any directory of RULE/SKILL/WORKFLOW files. Runs per-document audits plus the cross-document refactoring pass (split / join / merge / clone / reformat). Use when reviewing the structure of a documentation set, not just one file. For single-document audits, use audit-rule, audit-skill, or audit-workflow directly.

SKILL.md

4.7 KB, 972 tokens by cl100k_base, as published. Nobody here has run it

Audit Collection

Audit a body of related agent documents. Runs per-document audits and adds the cross-document refactoring pass.

When to Use

  • Reviewing a project's .kiro/ tree, ~/.claude/skills/ directory, or docs/agent/ collection
  • Reorganizing an organization's standard SOPs / agent rules
  • Migrating documents between vendors (Claude Code → Cursor, etc.) and want to consolidate
  • Onboarding a new agent doc collection from another project / person

When NOT to Use

  • Auditing one document → use audit-rule, audit-skill, or audit-workflow directly
  • Just looking up which checklist applies → read steering-audit-suite/shared/

Workflow

  1. Enumerate the documents in scope. List every file that prescribes agent behavior (RULE, SKILL, WORKFLOW), with paths.
  2. Classify each document by type. Use discriminators in steering-audit-suite/shared/doc-types.md. Note any document whose type doesn't match its file location/format (Reformat candidate).
  3. Per-document audit pass. For each document, invoke the appropriate single-doc audit (audit-rule, audit-skill, or audit-workflow). Collect findings.
  4. Cross-document refactoring pass. Apply the operations in steering-audit-suite/shared/refactoring.md:
    • Split: any document covering multiple unrelated concerns?
    • Join: any documents that always need to be read together?
    • Merge: any documents with overlapping or contradictory content?
    • Clone: any pattern that should also live elsewhere (e.g., promoted from project to global)?
    • Reformat: any document whose content is one type but its format/location declares another?
  5. Conflict and overlap analysis:
    • Do any two documents contradict each other?
    • Do any two documents share the same trigger condition (would both invoke for the same input)?
    • Are any documents orphaned (referenced nowhere, invoked never)?
  6. Vendor-mismatch analysis (if the collection spans vendors):
    • Are documents intended to apply across vendors duplicated correctly?
    • Are documents in vendor-specific formats but with vendor-agnostic content (clone candidates for other vendors)?
  7. Produce the consolidated audit report in the format defined in steering-audit-suite/shared/output-format.md. Per-doc findings + cross-doc refactor proposals.

Output Structure

# Collection Audit: <root path>

## Inventory
| Path | Type | Vendor | Status |
|---|---|---|---|

## Per-Document Findings
### <path>
<findings from audit-rule/skill/workflow>

(...)

## Cross-Document Findings
### Refactor: <operation>
<source, destination, reason>

### Conflicts
<list of contradictions or overlapping triggers>

### Orphans
<documents referenced nowhere or never invoked>

## Vendor Coverage
<which document patterns exist in which vendor formats; gaps>

Reference Files

  • steering-audit-suite/shared/doc-types.md
  • steering-audit-suite/shared/checklist.md
  • steering-audit-suite/shared/refactoring.md
  • steering-audit-suite/shared/portability.md
  • steering-audit-suite/shared/reference-exemplars.md
  • steering-audit-suite/shared/output-format.md

Anti-Patterns

  • Don't execute split/join/merge/clone/reformat operations without explicit authorization. Propose them. The author decides.
  • Don't propose vendor coverage for collections that are intentionally single-vendor.
  • Don't surface every per-document finding as a collection-level concern. Per-doc findings stay per-doc; the collection pass focuses on cross-document structure.
  • Anti-patterns and the audit checklist are non-exhaustive — see steering-audit-suite/shared/checklist.md and use the "Other Findings" section of the report for issues outside the enumerated checks.

Resilience to Correction

If the document owner contests a finding, quote the finding back with the owner's response in the audit record. Do not silently drop a contested finding; do not double down without engaging with the owner's reasoning.

Standalone-Mode Note

When this SKILL.md is copied out of the repo (e.g., to ~/.claude/skills/audit-collection/), the relative references to steering-audit-suite/shared/ will not resolve. In that case, inline the relevant shared content into the skill body, or maintain the suite as a checked-out directory the user can point at.

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.