/seo-technical

name: seo-technical
description: Technical SEO site audit via the DataForSEO On-Page API. Crawls the site (up to 100 pages by default), surfaces critical errors (broken links, redirect chains, missing meta, slow pages, schema gaps, indexability issues), and returns a Technical Score (0-100) with prioritized fixes.
allowed-tools:
  - Bash
  - Write

Phase 0: Credential Preflight (REQUIRED — run BEFORE anything else)

Before running any of the steps below, **always** invoke the shared preflight check:

~/.claude/skills/seo/scripts/preflight.sh

**If exit code is 0:** credentials are configured — proceed with the rest of this skill silently.

**If exit code is 2:** the script prints the DataForSEO setup wizard to stdout. STOP, display that wizard to the user verbatim, and **wait for them to paste credentials** in this format:

login: their_email@example.com
password: their_api_password_here

When they reply:

1. Parse `login:` and `password:` from their message. 2. Write them to `~/.claude/skills/seo/.env`:

   DATAFORSEO_LOGIN=<login>
   DATAFORSEO_PASSWORD=<password>

3. `chmod 600 ~/.claude/skills/seo/.env` 4. Run a verification call: `~/.claude/skills/seo/scripts/keyword_research.py volume "test"` 5. If verification succeeds (real JSON returned): tell the user "✅ Credentials verified. Running your command now..." and proceed with the original request. 6. If status `40104 — Please verify your account`: tell the user to verify their account at https://app.dataforseo.com/, then say "continue" to retry. 7. If any other auth error: ask them to double-check the API password (the long alphanumeric string from https://app.dataforseo.com/api-access — not their account login password).

**Never** echo credentials back to the user, never include them in tool output, and never commit them.

---

Technical SEO Audit Skill

> **Powered by:** [DataForSEO API](https://dataforseo.com) — On-Page `instant_pages` (single URL) and On-Page `task_post` + `summary` + `pages` (full crawl). > **Cost:** ~$0.05-0.10 per 100-page crawl. Crawl takes 2-5 minutes.

Run

For a single page (instant):

~/.claude/skills/seo/scripts/on_page_audit.py page --url <url>

For a full site crawl (2-5 min for 100 pages):

~/.claude/skills/seo/scripts/on_page_audit.py site --target <domain> --max-crawl-pages 100

Categories to evaluate

Pull from the `summary.checks` and per-page `checks` objects:

Indexability & Crawl

• `is_4xx_code`, `is_5xx_code`, `is_redirect` chains > 2 hops
• `canonical` mismatches
• `robots_txt`, blocked resources
• `is_https`, mixed content

On-page meta

• Missing/duplicate `<title>`, `<meta description>`
• Title length (should be 50-60 chars), description (140-160)
• `h1` missing or duplicate
• Heading hierarchy gaps

Speed & Core Web Vitals

• `time_to_interactive`, `largest_contentful_paint`
• `cumulative_layout_shift`
• Resource errors

Content quality

• `plain_text_word_count` < 300 = thin content
• `plain_text_rate` (text vs. HTML)
• `automated_readability_index`

Structured data

• Schema present? Which types?
• Missing organization / article / breadcrumb / product schema for the site type

Issue priority rubric

| Priority | Examples | |----------|----------| | **Critical** | 5xx errors, no-index on key pages, broken canonical, mixed content, robots.txt blocking | | **High** | Missing titles/descriptions on indexable pages, broken internal links, redirect chains, slow LCP (>4s) | | **Medium** | Duplicate titles/descriptions, missing alt text at scale, thin content (<300 words) on key pages | | **Low** | Heading hierarchy gaps, suboptimal title length, missing schema |

Technical Score (0-100)

Start at 100. Subtract:

• 15 per critical issue (max 60)
• 5 per high issue (max 30)
• 1 per medium issue (max 15)
• 0.2 per low issue (max 5)

Floor at 0.

Return JSON shape

{
  "technical_score": 72,
  "crawl_summary": {"pages_crawled": 100, "errors": 8, "warnings": 23},
  "critical_issues": [{"title": "...", "count": 3, "recommendation": "..."}],
  "high_issues": [...],
  "medium_issues": [...]
}