skills.rest
nyosegawa
nyosegawaCOMMUNITY
Explore all skills by this author →
𝕏@gyakuseGitHub

skill-auditor

Parse session transcripts and skill manifests to generate an interactive HTML report.

56|Updated Feb 11, 2026
Search Tags:#code quality#portfolio optimization#skill audit#routing analysis#LLM agent evaluation
One-click install
npx skills add https://github.com/nyosegawa/skills --skill skill-auditor-nyosegawa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-auditor
Source: https://github.com/nyosegawa/skills/tree/main/skills/skill-auditor
Command: npx skills add https://github.com/nyosegawa/skills --skill skill-auditor-nyosegawa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires tiktoken, and includes scripts (resource) and references (resource) and agents (resource) components.

What problem does it solve?

This Skill identifies and fixes issues in your entire skill portfolio, ensuring they work together effectively and accurately route user requests.

Core Features & Use Cases

  • Portfolio Health Analysis: Evaluates routing accuracy, attention competition, and coverage gaps across all your skills.
  • Actionable Patches: Generates specific, safe edits to skill descriptions to improve routing.
  • Use Case: After deploying several new skills, you notice users aren't getting the results they expect. Run skill-auditor to pinpoint which skills are competing, which are missing intents, and get concrete suggestions to fix them.

Quick Start

Use the skill-auditor to analyze your current skill set and generate an HTML report.

Frequently Asked Questions about skill-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit LLM agent routing performance and skill coverage gaps?▼

Analyze session transcripts and skill manifests to identify routing errors, attention competition, and coverage gaps, generating an interactive HTML report with actionable patches and new skill suggestions.

What is the best way to fix Claude Code skill routing errors after deploying new skills?▼

Pinpoint competing skills and missing intents across your portfolio, then generate specific, safe edits to skill descriptions to improve routing accuracy and resolve attention competition.

Do I need external API keys to parse session transcripts for skill portfolio analysis?▼

No external API keys are required for skill portfolio analysis. The process uses Claude sub-agents internally to evaluate routing accuracy, attention competition, and coverage gaps directly from session transcripts.

Can I use sub-agents to evaluate code quality and skill routing without external dependencies?▼

Yes, Claude sub-agents evaluate skill routing and code quality without external API dependencies, requiring only the tiktoken library to parse manifests and session transcripts for coverage gaps.

What are the limitations of using sub-agents for LLM agent evaluation?▼

Limitations of sub-agent LLM agent evaluation include relying solely on the tiktoken dependency and requiring existing session transcripts and skill manifests, meaning it cannot analyze routing performance for unlogged sessions.

Related Skills

Explore More Software Engineering Skills→
mindmorassmindmorass
2

video-upload-patterns

Automate video uploads to YouTube, TikTok, and Vimeo with metadata and scheduling.

Community
Advanced
yAtomtomyAtomtom

skill-auditor

Analyze Claude Code session transcripts to assess skill portfolio health and routing accuracy.

Community
Advanced
furedeafuredea
1

skill-auditor

Audit Claude Code or Codex skill portfolios for routing errors and coverage gaps.

Community
Advanced
DhumitechDhumitech

agent-evaluation

Automate AI agent evaluation and benchmarking with repeatable test pipelines.

Community
Advanced