workspace-audit

Audit workspace integrity, formatting, dependencies, rules, token budgets, and project health.

4|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/CaufieldZ/pm-workspace-public --skill workspace-audit-caufieldz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: workspace-audit
Source: https://github.com/CaufieldZ/pm-workspace-public/tree/main/.claude/skills/workspace-audit
Command: npx skills add https://github.com/CaufieldZ/pm-workspace-public --skill workspace-audit-caufieldz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Audit the entire workspace to identify and report issues across file integrity, formatting, dependencies, rules, token budgets, and project health.

Core Features & Use Cases

  • Two-phase audit: Phase 1 Bash script for hard checks and Phase 2 model-driven semantic review, enabling both deterministic validation and nuanced analysis.
  • Category-scoped auditing: selectively run specific audit categories to balance speed and coverage across projects.
  • Deliverables: concise in-chat report plus a detailed, evidence-rich file with actionable remediation guidance.

Quick Start

Run bash .claude/skills/workspace-audit/references/audit.sh with selected categories to start auditing.

Frequently Asked Questions about workspace-audit

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

FAQPage Schema
How do I run a workspace audit to check file integrity and frontmatter formatting?

To audit workspace quality assurance, run the provided Bash script with selected categories to perform deterministic hard checks, then trigger a model-driven semantic review for deeper analysis and actionable remediation.

What is the best way to check project dependencies and token budgets?

A category-scoped workspace audit validates project dependencies and token budgets by running targeted Bash hard checks and semantic reviews, producing a detailed deliverable with per-item evidence.

Can I selectively audit specific categories to speed up the workspace review?

Category-scoped auditing supports selectively running specific audit categories to balance speed and coverage, ensuring targeted validation of frontmatter, references, and project health without a full scan.

How does the two-phase workspace audit mechanism work?

Phase 1 executes a deterministic Bash script for hard validation checks, while Phase 2 applies model-driven semantic review to analyze workspace quality, yielding both an in-chat report and a detailed remediation file.

What deliverables do I get after running a project health audit?

The workspace audit produces two deliverables: a concise in-chat report for quick overview and a detailed file with per-item evidence, ensuring traceability and actionable remediation for identified formatting and dependency issues.

Why does my workspace audit need both a Bash script and a semantic review?

The Bash script provides deterministic hard checks for file integrity and dependencies, whereas the model-driven semantic review handles nuanced analysis of rules and frontmatter, together ensuring comprehensive workspace quality assurance.