← ClaudeAtlas

capability-auditorlisted

Audit capability artifacts for static safety, evidence coverage, duplication, context cost, prompt contracts, dependencies, and install risk. Use capability-evaluation for behavioral baselines and portfolio architect for structural change.
Xopoko/plug-n-skills · ★ 11 · AI & Automation · score 77
Install: claude install-skill Xopoko/plug-n-skills
# Capability Auditor Bundled commands use `$PLUGIN_ROOT` (`$env:PLUGIN_ROOT` in PowerShell; same path suffix) for the plugin root. Set it once: use the host's plugin-root variable when defined (Claude Code: `PLUGIN_ROOT="$CLAUDE_PLUGIN_ROOT"`), otherwise the absolute path of this skill folder's `../..`. Use this for independent review before adopting, installing, publishing, or relying on a skill or plugin. This is a static/source audit. When the decision depends on whether a candidate artifact changes agent behavior relative to an explicit baseline, route the controlled comparison and adoption receipt to `capability-evaluation`. Route runner, orchestration, cancellation, recovery, or harness reliability failures to Agent Harness. ## Audit Spine 1. Identify the subject: skill folder, plugin folder, marketplace entry, synthesis package, or candidate source. 2. Inventory files and metadata. 3. Classify risks as `required`, `optional`, `example`, `advisory`, or `hidden`. 4. Measure context density when the subject affects hot context or repeated agent behavior. 5. If audit findings imply split, merge, delete, router, reference-extract, or script-extract decisions, hand off to `capability-portfolio-architect` before changing structure. 6. Validate structure and run deterministic self-tests. 7. For synthesized or installed outputs, validate install scope, source surface, install requirement, and whether any required activation proof exists. 8. Produce a verdict: eligible, eli