agent-audit
Audit agents spawned in the current/last run against the agent-selection taxonomy
Run quality-scores + token-budget + dep-graph, interpret results, flag missing budgets, orphans, and oversize skills
> /plugin marketplace add heymegabyte/claude-skillsHow it fires
How this command gets triggered: by you, by Claude, or both.
/skill-healthContext preview
What this command does when you run it.
Run quality-scores + token-budget + dep-graph, interpret results, flag missing budgets, orphans, and oversize skills
description: Run quality-scores + token-budget + dep-graph, interpret results, flag missing budgets, orphans, and oversize skills argument-hint: [optional pack filter]
Run the full skill health suite against the plugin's SKILL.md files.
**Purpose** — catch structural decay: skills without descriptions, budgets, dependencies, or pack membership.
**When to use** — `/skill-health` after any skill add/edit; weekly audit; before release.
**Inputs** — `$ARGUMENTS` (optional pack name to scope). Runs all three tools in sequence.
1. **Quality scores** — `node bin/skill-quality-scores.mjs [--threshold <12>]` — flag skills scoring ≤12/18.
2. **Token budgets** — `node bin/skill-token-budget.mjs [--pack $ARGUMENTS]` — total budget and slot-level breakdown.
3. **Dependency graph** — `node bin/skill-dep-graph.mjs [--pack $ARGUMENTS]` — adjacency and pack density.
For each finding, produce one bullet with skill · problem · fix path.
**Verification** — re-run `/skill-health` after applying fixes; zero findings = pass.
**Can update** `_packs/*.yml`, SKILL.md frontmatter, or skill body (add budget markers, descriptions, pack assignment).
14-category autonomous product-building OS for 32+ AI coding tools. One-line prompts → deployed products.
Repo: heymegabyte/claude-skills
Audit agents spawned in the current/last run against the agent-selection taxonomy
Run the Agent Diversity Review gate and emit the result table
Meta-analyze the effectiveness of a /loop arc — per-iteration metrics, LOC delta trend, saturation detection, and a keep/lengthen/delete recommendation.
Audit the rules/ directory for missing foundational principles; output gap list with priority and justification
Validate ~/.claude/settings.json hooks block — event names, file existence, executability, matcher syntax; --fix repairs common issues
Catch Resend-class bug (isError: false on HTTP 4xx/5xx) across all MCP server tool handlers