seo-auditor

Run the SEO auditor on documentation files. Target path: $ARGUMENTS (default: all docs/ and root README.md).

If $ARGUMENTS is --report-only, scan without making changes.

Execute all 7 phases. Auto-fix non-destructive issues. Never change URLs. Preserve content on high-ranking pages.

Phase 1: Discovery

Find all target markdown files:

• docs/**/*.md — all documentation pages
• README.md files in domain root directories
• If $ARGUMENTS specifies a path, scope to that path only

For each file, extract current state: title: frontmatter, description: frontmatter, H1, H2s, word count, link count. Store as baseline for the report.

Identify recently changed files: git log --oneline -2 --name-only -- docs/ README.md

Phase 2: Meta Tags

For each file with YAML frontmatter:

Title (title: field):

• Must be 50-60 characters
• Must contain a primary keyword
• Must be unique across all pages
• Auto-fix generic titles using domain context

Description (description: field):

• Must be 120-160 characters
• Must contain primary keyword
• Must be unique — no duplicates
• Auto-fix from SKILL.md frontmatter or first paragraph

Run SEO checker on built HTML pages:

python3 marketing-skill/seo-audit/scripts/seo_checker.py --file site/{path}/index.html

Phase 3: Content Quality

Heading structure: One H1 per page, no skipped levels, keywords in headings.

Readability: Run content scorer:

python3 marketing-skill/content-production/scripts/content_scorer.py {file}

Target: readability ≥ 70, structure ≥ 60.

AI detection (on non-generated files only):

python3 marketing-skill/content-humanizer/scripts/humanizer_scorer.py {file}

Flag pages < 50. Fix AI clichés: "delve", "leverage", "it's important to note", "comprehensive".

Do NOT rewrite pages ranking well — only fix critical issues on those.

Phase 4: Keywords

Check each page has its primary keyword in: title, description, H1, first paragraph, at least one H2.