One-click install
npx skills add https://github.com/semicolon-devteam/semo --skill audit-semo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit-semo
Source: https://github.com/semicolon-devteam/semo/tree/main/packages/meta/skills/audit-semo
Command: npx skills add https://github.com/semicolon-devteam/semo --skill audit-semo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Bash, Read, Grep, Glob, and includes references (resource) components.

What problem does it solve?

This Skill performs a comprehensive quality audit across all SEMO packages, automatically detecting standard violations, inefficient structures, and quality issues in Agents, Skills, and Commands, ensuring consistency and maintainability.

Core Features & Use Cases

  • Agent Validation: Checks for mandatory frontmatter fields (name, description, tools, model), "PROACTIVELY use when" patterns, and line count limits.
  • Skill Validation: Verifies frontmatter, mandatory system messages, "Use when" patterns, and line count limits.
  • Command Validation: Confirms file existence, frontmatter, and CLAUDE.md integration.
  • Severity-Based Reporting: Categorizes issues as Critical, Important, or Nice-to-have with actionable recommendations.
  • Use Case: To ensure all SEMO components adhere to team standards, you can run "전체 SEMO 패키지 통합 품질 감사" to get a detailed report on any issues and how to fix them.

Quick Start

Audit all SEMO packages: "전체 패키지 품질 점검" Check Agent/Skill standards: "Agent/Skill/Command 표준 준수 검토" Detect inefficient structures: "비효율적 구조 탐지"

Frequently Asked Questions about audit-semo

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

FAQPage Schema
How do I audit SEMO packages for quality and standard violations?

Run a comprehensive quality audit across all SEMO packages to detect standard violations and quality issues in Agents, Skills, and Commands. The audit checks frontmatter fields, line-count thresholds, system messages, and cross-package reporting to surface nonconformities with severity-based categorization.

What standards does code audit enforce for Agents and Skills?

Code audit enforces mandatory frontmatter fields, model values (opus, sonnet, haiku, inherit), line-count limits (Agents up to 200 lines, Skills up to 100 lines), system message presence, and "Use when" patterns to ensure consistency across all SEMO components.

Can I validate Command files and their CLAUDE.md linkage?

Yes. Command validation confirms file existence, verifies frontmatter structure, and checks CLAUDE.md integration to ensure all Commands meet team standards and maintain proper documentation linkage.

How are audit findings reported and prioritized?

Findings are categorized by severity—Critical, Important, or Nice-to-have—with actionable recommendations. The structured audit output spans all SEMO packages (semo-meta, semo-next, semo-backend, semo-po, semo-qa, semo-core, semo-pm) for complete visibility.

What dependencies does a code audit tool require?

Code audit uses Bash, Read, Grep, and Glob to inspect package contents and extract frontmatter patterns. These dependencies enable efficient pattern matching and file traversal across multiple SEMO packages.

Does this tool work with existing Agent and Skill implementations?

Yes. It audits existing implementations by inspecting Agent frontmatter, PROACTIVELY use when patterns, Skill frontmatter, system messages, and Progressive Disclosure structures to identify gaps without requiring code modifications.