skills.rest
greglas75
greglas75COMMUNITY
Explore all skills by this author →
GitHub

content-fix

Applies safety-tiered fixes from content-audit JSON findings to content files.

6|Updated Mar 27, 2026
Search Tags:#typography#auto-fix#encoding#mojibake#content audit#cms migration#markdown fix
One-click install
npx skills add https://github.com/greglas75/zuvo --skill content-fix-greglas75
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: content-fix
Source: https://github.com/greglas75/zuvo/tree/main/skills/content-fix
Command: npx skills add https://github.com/greglas75/zuvo --skill content-fix-greglas75

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After running a content audit, teams face dozens of mechanical issues like encoding artifacts, broken markdown, and CMS migration debris that are tedious and error-prone to fix by hand. This Skill reads the audit JSON and applies the safe fixes automatically with verification and rollback. ## Core Features & Use Cases - Safety-tiered fix application: Classifies findings as SAFE, MODERATE, or MANUAL and applies only what the selected mode allows (default, --auto, --dry-run). - Encoding and markdown repair: Strips NBSP, zero-width characters, BOM, and mojibake; closes unclosed italics; removes WordPress shortcodes, PHP tags, and font tags. - Guardrails and verification: Dirty-file checks, stale-audit protection, build verification with per-file rollback, adversarial review, and a JSON fix report. - Use Case: After a content-audit flags 42 NBSP characters and mojibake across migrated blog posts, run this Skill to fix the SAFE findings automatically and generate a report of items needing manual review. ## Quick Start Run the content-fix skill on the latest content-audit JSON to apply all SAFE fixes and report what remains for manual review.

Frequently Asked Questions about content-fix

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

FAQPage Schema
How do I automatically fix content-audit findings?▼

Run the content-fix skill after a content-audit produces a JSON report. By default it applies only SAFE fixes like encoding-strip, markdown-fix, and typography-fix; use --auto to also apply MODERATE fixes such as mojibake repair and artifact removal.

What is the difference between SAFE and MODERATE fixes?▼

SAFE fixes (encoding-strip, markdown-fix, typography-fix) are applied in default mode and carry minimal risk. MODERATE fixes (encoding-mojibake, artifact-remove) require the --auto flag because they alter content more aggressively, such as replacing garbled characters or removing shortcodes.

Can I preview content fixes before applying them?▼

Yes, use the --dry-run flag to show exactly what would be fixed without changing any files. You can also target specific findings with --finding or specific categories with --fix-type.

What happens if a content fix breaks the build?▼

The skill runs the project build command after applying fixes. If the build fails, it rolls back the affected files using in-memory snapshots taken before modification, and reports the failure.

Why are some findings marked NEEDS_REVIEW and not fixed?▼

Findings are marked NEEDS_REVIEW when the target file has uncommitted changes, or when a fix is inherently ambiguous, such as Joomla image paths whose correct target is unknown. These always require human review even in --auto mode.

Related Skills

Explore More Content & Communication Skills→
gaiastudio-aigaiastudio-ai
2

gaia-review-a11y

Identify WCAG 2.1 accessibility issues and generate a Markdown findings report.

Official
Advanced
Corvalis-LLCCorvalis-LLC
6

codex-validation

Coordinate end-to-end project validation with a findings-first manual review.

Official
Advanced
CodeAlive-AICodeAlive-AI
3

seo-audit

Diagnose SEO issues on websites and identify prioritized fixes.

Official
Advanced
getsentrygetsentry
44.7k

lint-fix

Fix eslintPluginScraps rule violations across the Sentry codebase using autofix, batching, and codemods.

Official
Intermediate