Helpful Content & E-E-A-T Quality Auditor

Overview

Audit technical articles for Google Helpful Content guidelines, technical SEO, GEO/AEO answer extractability, factual accuracy, and publishing readiness.

Verification & Test Notes

[!NOTE] Tested & Reviewed on 2026-10-10 for runtime environment Google HCS Guidelines, Markdown AST.

Runs comprehensive pre-publish audits. Catches passive voice, thin generic claims, missing benchmarks, and heading gaps.

Installation & Setup

Option 1: Direct Skill Download

1. Download Helpful Content & E-E-A-T Quality Auditor (content-quality-auditor.zip).
2. Unpack the zip file into your workspace agent directory:
   • Antigravity IDE: .agents/skills/content-quality-auditor/
   • Claude Code: ~/.claude/skills/content-quality-auditor/ or .agents/skills/content-quality-auditor/
   • Cursor: .cursor/rules/content-quality-auditor/ or your project prompt definitions.

Option 2: CLI Activation

If using Antigravity or compatible agent runners:

# Load into active workspace
mkdir -p .agents/skills/content-quality-auditor
curl -sL https://nadhebe.com/skills/content-quality-auditor.zip -o /tmp/content-quality-auditor.zip
unzip /tmp/content-quality-auditor.zip -d .agents/skills/content-quality-auditor/

Skill Instructions & Reference

Content Quality & SEO/AEO Auditor

Activate this skill to audit, score, and verify technical articles prior to publishing.

1. When to Activate This Skill

• Before publishing a newly drafted article or guide.
• When evaluating existing content for Google Helpful Content and EEAT compliance.
• When auditing markdown files for formatting errors, broken links, or frontmatter schema violations.
• When checking answer extractability for Google AI Overviews, Perplexity, and ChatGPT.

2. Five Audit Pillars

Pillar 1: Factual Integrity & Zero Fabrication (Strict)

• Zero Hallucinated Citations: Every URL cited in sources must be live and official.
• No Fabricated Benchmarks: Model numbers, token limits, and prices must reflect official documentation.
• No Fake Quotes / Experiences: Avoid fabricated anecdotes (“In my 10 years using Claude Code…”). Frame advice through objective engineering principles.
• Zero AI Clichés: Ban filler words like “In today’s fast-paced digital landscape”, “delve into”, “testament to”, “game changer”, “seamlessly”.

Pillar 2: Technical SEO & Heading Hierarchy

• Single H1: Exactly one # Title per article.
• Strict Hierarchy: H1 → H2 → H3. Never skip levels (no H1 → H3).
• Title Formula: Primary Keyword First + Benefit + | Nadhebe.
• Meta Description: 150–160 characters, containing the primary keyword.

Pillar 3: GEO / AEO Answer Extractability

• Direct Answer Blocks: 2–3 sentence direct answer immediately below each H2 heading.
• Data Tables: Comparisons must use structured markdown tables, not bulleted paragraphs.
• Self-Contained Definitions: Sections must be extractable by an LLM without reading the entire page.

Pillar 4: Astro Schema & Frontmatter Validation

Verify all required frontmatter fields for src/content/config.ts:

• title, description, pubDate, author: nadhebe-team, category, tags
• heroImage (must exist in /public/images/<slug>-hero.webp)
• topic, isPillar, searchIntent
• faq: minimum 3-4 structured Q&A pairs
• sources: minimum 2 authoritative links

Pillar 5: Link Integrity & Build Safety

• All internal markdown links must point to valid routes (e.g. [Text](/guides/slug)).
• Must verify that npm run build compiles with 0 broken links.

3. Audit Scoring Template

# Content Quality Audit Report: [Article Title]

| Dimension | Score (1-100) | Status | Key Findings |
| :--- | :--- | :--- | :--- |
| **Factual Integrity & EEAT** | [Score] | ✅ Pass / ❌ Fail | [Details] |
| **Technical SEO & Structure**| [Score] | ✅ Pass / ❌ Fail | [Details] |
| **GEO / AEO Extractability** | [Score] | ✅ Pass / ⚠️ Warn | [Details] |
| **Frontmatter & Schema** | [Score] | ✅ Pass / ❌ Fail | [Details] |
| **Link Integrity & Assets** | [Score] | ✅ Pass / ❌ Fail | [Details] |

### Summary & Verdict
- **Overall Quality Score**: [Score] / 100
- **Publish Readiness**: [Ready to Publish / Revisions Needed]

### Actionable Fixes (if any)
1. **[Fix 1]**: [Exact modification required]
2. **[Fix 2]**: [Exact modification required]