content-audit

Audits markdown content files for encoding artifacts, broken links, and CMS migration debris.

6|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/greglas75/zuvo --skill content-audit-greglas75
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: content-audit
Source: https://github.com/greglas75/zuvo/tree/main/skills/content-audit
Command: npx skills add https://github.com/greglas75/zuvo --skill content-audit-greglas75

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Content migrated from legacy CMS platforms or edited across teams accumulates invisible encoding corruption, broken markdown syntax, dead links, and migration leftovers that silently degrade site quality. This Skill scans content files across 8 dimensions (CC1-CC8) and produces a graded report with auto-fixable findings tagged for a companion fix skill. ## Core Features & Use Cases - Encoding and mojibake detection: Finds NBSP, zero-width characters, BOM, and multi-encoding mojibake signatures (Latin-1, ISO-8859-2, Windows-1252) with file:line evidence. - Link and image integrity: Resolves internal links and image paths against the file tree, validates anchor fragments against headings, and optionally probes live URLs for 404s and redirect chains. - CMS migration cleanup: Detects Joomla paths, WordPress shortcodes, PHP tags, WYSIWYG inline styles, and unexpanded template variables left over from migrations. - Use Case: After migrating a blog from WordPress to Astro, run the audit to surface 25 findings including mojibake in 4 posts and 8 broken internal links, then hand the auto-fixable items to the content-fix skill. ## Quick Start Ask the AI to run a content audit on the content directory of your site and report the grade with all blocking findings.

Frequently Asked Questions about content-audit

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

FAQPage Schema
How do I audit markdown content files for quality issues?

Run the content audit against your content directory to scan 49 checks across 8 dimensions covering encoding, markdown syntax, migration artifacts, frontmatter, links, images, completeness, and typography. Results are graded A through D with file:line evidence for every finding.

How to detect mojibake and encoding corruption in content files?

The audit greps for mojibake signatures from Latin-1, Windows-1252, ISO-8859-2, and Turkish encodings, plus NBSP, zero-width characters, BOM markers, and replacement characters. Mojibake detection is a blocking check that caps the grade at D.

Does the content audit work with Hugo, Astro, and Next.js sites?

Yes, it auto-detects the framework from config files and applies the matching content path conventions, such as content/ for Hugo and src/content/ for Astro. Template syntax in .astro, .tsx, and .mdx files is stripped before text checks to avoid false positives.

Can the audit check external links and live image URLs?

Yes, pass --live-url with a base URL or --check-external to enable HTTP HEAD checks for dead links, image 404s, and redirect chains. Without these flags, live-dependent checks report INSUFFICIENT DATA rather than failing.

What are the limitations of the quick audit mode?

Quick mode runs only grep-based source checks for encoding, migration artifacts, and path resolution without dispatching agents. Frontmatter parsing, content completeness, and spelling checks are skipped and reported as INSUFFICIENT DATA.

Does the content audit modify my files?

No, the audit is strictly read-only against source files. It writes reports only to the zuvo/audits/ directory, and fixes are delegated to the separate content-fix skill for findings tagged with a fix_type.