skill-stocktake

Audit Claude skills and commands with a repeatable stocktake workflow.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/shimo4228/claude-code-learned-skills --skill skill-stocktake
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-stocktake
Source: https://github.com/shimo4228/claude-code-learned-skills/tree/main/skills/skill-stocktake
Command: npx skills add https://github.com/shimo4228/claude-code-learned-skills --skill skill-stocktake

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, and includes scripts (resource) components.

What problem does it solve?

Auditing Claude skills and commands for quality in a repeatable, scalable workflow to ensure currency and consistency across projects.

Core Features & Use Cases

  • Inventory and Quick Scan vs Full Stocktake modes to limit effort while preserving coverage.
  • Phase-driven evaluation using a structured checklist and UTC timestamped results.
  • Project-wide and global skill coverage, with diff-based incremental updates and resumable runs.

Quick Start

Run the stocktake command from a project root to perform a Quick Scan, or pass the full flag to perform a complete stocktake.

Frequently Asked Questions about skill-stocktake

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

FAQPage Schema
How do I audit Claude skills for quality and consistency?

To audit Claude skills, run the stocktake command from your project root to evaluate metadata and enforce a structured checklist, generating a UTC timestamped results.json manifest for quality tracking.

What is the difference between a Quick Scan and a Full Stocktake?

A Quick Scan evaluates only changed skills using diff-based incremental updates to limit effort, while a Full Stocktake performs a comprehensive review of all global and project-scoped skills for complete coverage.

Do I need jq to run a skill stocktake?

Yes, jq is required as a dependency to execute the skill stocktake workflow, parse frontmatter-driven metadata models, and process the structured evaluation checklist outputs.

How does the stocktake workflow handle project-scoped versus global skills?

The stocktake workflow targets both global and project-scoped skill directories by inventorying them together, allowing you to audit Claude commands across multiple scopes while maintaining consistent metadata standards.

Can I resume an interrupted Claude skill audit?

Yes, the stocktake workflow supports resumable runs, allowing you to pause and continue a comprehensive skill audit without losing previous UTC timestamped results or evaluation progress.

What metadata format does the skill audit enforce?

The audit enforces a frontmatter-driven metadata model, requiring skills to contain structured frontmatter fields that are validated against a structured evaluation checklist during the stocktake process.