solo-seo-audit

# /seo-audit

SEO health check for any URL or project landing page. Fetches the page, analyzes meta tags, OG, JSON-LD, sitemap, robots.txt, checks SERP positions for target keywords, and outputs a scored report.

## MCP Tools (use if available)

- `web_search(query, engines, include_raw_content)` — SERP position check, competitor analysis
- `project_info(name)` — get project URL if auditing by project name

If MCP tools are not available, use Claude WebSearch/WebFetch as fallback.

## Steps

1. **Parse target** from `$ARGUMENTS`.
   - If URL (starts with `http`): use directly.
   - If project name: look up URL from project README, CLAUDE.md, or `docs/prd.md`.
   - If empty: ask via AskUserQuestion — "Which URL or project to audit?"

2. **Fetch the page** via WebFetch. Extract:
   - `<title>` tag (length check: 50-60 chars ideal)
   - `<meta name="description">` (length check: 150-160 chars ideal)
   - Open Graph tags: `og:title`, `og:description`, `og:image`, `og:url`, `og:type`
   - Twitter Card tags: `twitter:card`, `twitter:title`, `twitter:image`
   - JSON-LD structured data (`<script type="application/ld+json">`)
   - `<link rel="canonical">` — canonical URL
   - `<html lang="...">` — language tag
   - `<link rel="alternate" hreflang="...">` — i18n tags
   - Heading structure: H1 count (should be exactly 1), H2-H3 hierarchy

3. **Check infrastructure files:**
   - Fetch `{origin}/sitemap.xml` — exists? Valid XML? Page count?
   - Fetch `{origin}/robots.txt` — exists? Disallow rules? Sitemap reference?
   - Fetch `{origin}/favicon.ico` — exists?

4. **Forced reasoning — assess before scoring:**
   Write out before proceeding:
   - **What's present:** [list of found elements]
   - **What's missing:** [list of absent elements]
   - **Critical issues:** [anything that blocks indexing or sharing]

5. **SERP position check** — for 3-5 keywords:
   - Extract keywords from page title + meta description + H1.
   - For each keyword, search via MCP `web_search(query="{key...