seo-audit-full — Advanced Full SEO Audit

This skill runs a full single-page SEO audit from the seo-audit-full directory. It does not route to seo-audit when only a URL is provided.

When to Use This Skill

Use seo-audit-full when the user asks for:

• "seo-audit-full"
• "full SEO audit"
• "advanced SEO audit"
• "technical SEO audit"
• "deep audit"
• "comprehensive SEO review"
• "audit everything"

URL-only requests are valid. When external datasets are not supplied, run the full public-signal workflow and clearly note missing data sources in the report.

Input Expected

At the start of a full audit, ask the user for a PageSpeed Insights API key:

For PageSpeed checks, please provide a Google PageSpeed Insights API key.
Get one here: https://developers.google.com/speed/docs/insights/v5/get-started
Open "Acquiring and using an API key" → "Get a Key".
If you do not provide one, I will stop before running the full audit because
PageSpeed is a required full-audit module.

Do not run seo-audit-full PageSpeed checks without a PageSpeed API key supplied by --api-key, PAGESPEED_API_KEY, or GOOGLE_PAGESPEED_API_KEY. If the key is missing, stop and ask the user to configure it instead of rendering a full report with missing PageSpeed data.

Architecture: Full = Core + Performance + Advanced

┌─────────────────────────────────────────────────────────────┐
│  seo-audit-full Workflow                                    │
│                                                             │
│  Phase 1: Run core scripts (./scripts/)                     │
│  ┌──────────────────────────────────────────────────────┐   │
│  │  check-site.py      → robots.txt, sitemap, 404, URL  │   │
│  │  check-page.py      → title, H1, meta desc, slug     │   │
│  │  check-schema.py    → JSON-LD validation              │   │
│  │  fetch-page.py      → raw HTML for analysis           │   │
│  └──────────────────────────────────────────────────────┘   │
│                          ↓                                  │
│  Phase 2: Run performance + full-only scripts (./scripts/)  │
│  ┌──────────────────────────────────────────────────────┐   │
│  │  check-pagespeed.py → Lighthouse scores + metrics     │   │
│  │  check-social.py    → OG Tags + Twitter Card          │   │
│  │  (more scripts added here as modules grow)            │   │
│  └──────────────────────────────────────────────────────┘   │
│                          ↓                                  │
│  Phase 3: LLM-only advanced checks                         │
│  ┌──────────────────────────────────────────────────────┐   │
│  │  E-E-A-T content quality scoring                      │   │
│  │  Duplicate content signals                            │   │
│  │  Anchor text quality assessment                       │   │
│  └──────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────┘

Output

Produce an Advanced Full SEO Audit Report by filling the template at assets/report-template.html, then save it to a file — never print raw HTML to the terminal.

File naming: reports/<hostname>-<slug>-full-audit.html

https://example.com/blog/best-tools → reports/example-com-blog-best-tools-full-audit.html
https://example.com/                → reports/example-com-full-audit.html

After saving, tell the user:

✅ Full Report saved → reports/example-com-full-audit.html
   Open it now? (yes / no)

If yes → run: open reports/example-com-full-audit.html

Template placeholders — fill each independently:

Scripts

Run full scripts from this directory. All output is structured JSON — use it directly as evidence.

Dependencies: pip install requests

Phase 1: Core scripts

# 1. site-level checks (robots.txt + sitemap.xml + 404 + URL canonicalization)
python scripts/check-site.py https://example.com

# 2. page-level checks (H1, title, meta description, canonical, URL slug)
python scripts/check-page.py https://example.com --keyword "primary keyword"

# 3. fetch raw HTML for downstream scripts
python scripts/fetch-page.py https://example.com --output /tmp/page.html

# 4. JSON-LD schema validation
python scripts/check-schema.py --file /tmp/page.html

Phase 2: Full-only scripts

# 5. PageSpeed / Lighthouse checks
python scripts/check-pagespeed.py https://example.com --strategy mobile --timeout 180 --api-key "USER_PROVIDED_KEY"
# If no key was provided, do not run this command. Ask the user to configure a PageSpeed API key first.

# 6. Social tags: OG + Twitter Card validation
python scripts/check-social.py --file /tmp/page.html
# Or directly from URL:
python scripts/check-social.py https://example.com

Each script exits with code 0 (all pass/warn) or 1 (any fail/error).

PageSpeed can take 200 seconds. Use a 180-second timeout by default. If it still times out, mark the Page Speed module as error and state that the PageSpeed API timed out; do not treat that as confirmed page performance failure. If PageSpeed fails because Google returns a quota/API-key error after a key was provided, keep the audit running and render the PageSpeed module as error with this instruction: Get a PageSpeed API key at https://developers.google.com/speed/docs/insights/v5/get-started → "Acquiring and using an API key" → "Get a Key".

Scope — Full Audit Check Whitelist

Full runs its own core checks plus the full-only items marked ★ below.

Site-Level Checks (in {{site_checks_html}})

Core checks:

• Sitemap URL Inventory · Staging Subdomain Indexation · robots.txt · sitemap.xml · 404 Handling · URL Canonicalization · i18n / hreflang

Staging Subdomain Indexation rules:

• Check common staging/test hosts before robots.txt: test., staging., dev., preview., beta., uat.
• Fail when a staging/test subdomain is publicly accessible, closely mirrors the production site, and is not protected by authentication, noindex, or a blocking robots.txt.
• Warn when a staging/test subdomain is publicly accessible but similarity or index protection cannot be confirmed.
• Pass when no public staging/test subdomain is detected, or detected staging hosts are protected by authentication, non-200 access, noindex, or Disallow: /.
• Explain the impact as duplicate indexable pages: Google may treat test.example.com and www.example.com as separate but near-identical URLs, splitting ranking signals and competing with the production site.
• Recommended fixes: add Basic Auth/password protection first; also block crawlers on the staging host with User-agent: * + Disallow: /; add page-level noindex if pages can still be accessed.

Sitemap URL Inventory rules:

• Place this as the first module in {{site_checks_html}}, before the Crawlability table.
• Render it as its own table with columns: Directory · URL Count · Page Type · Example Page.
• Use check-site.py sitemap_inventory output to summarize first-level directories, URL counts, inferred page types, and one representative example URL.
• Treat this as a site-level map, not a pass/fail single-page SEO audit.
• Use status info unless sitemap URLs cannot be parsed; do not penalize a site for having many or few URLs in a directory without deeper evidence.
• Always include a next-step note: the user can continue with deeper full audits by selecting representative sample URLs from major directories such as /blog/, /tools/, /alternatives/, /templates/, or /use-cases/.

Page Speed Checks (in {{pagespeed_checks_html}})

Full-only:

• Lighthouse category scores: Performance · Accessibility · Best Practices · SEO
• Lab metrics: FCP · LCP · TBT · CLS · Speed Index
• Final URL and screenshot availability

E-E-A-T Checks (in {{eeat_checks_html}})

Core checks:

• About Us · Contact · Privacy Policy · Terms of Service · Media/Partners (only if present)

Contact logic (Contact row only):

• A dedicated /contact page is not required
• Pass if contact is reachable via any of: dedicated contact page (HTTP 200) · About page with contact details · footer/nav mailto, email, social links, or contact form
• Fail only when no contact pathway exists anywhere on the site
• Missing /contact alone is not a fail when About or footer/nav already expose contact info

E-E-A-T infrastructure rules — two layers per trust page:

• Layer 1 — Exists: HTTP 200 for the trust page URL (Contact uses Contact-specific rules above)
• Layer 2 — Reachable: linked from footer or main nav

Status rules (About, Privacy, Terms, Media/Partners):

• Page missing (non-200) → Fail
• Page exists but not linked in footer/nav → Warn
• Page exists and linked in footer/nav → Pass
• Optional page missing → skip, do not include row

Contact-specific rules:

1. Try common paths (/contact, /contact-us) — HTTP 200 counts as Exists
2. If no contact page, check the About page body for email, social, or contact details
3. Also scan homepage footer and nav for mailto:, visible email, social links, or a contact form
4. Exists Pass if any contact pathway is found
5. Exists Fail only when no contact information is found anywhere
6. Reachable Pass if a contact page link or contact details appear in footer/nav
7. Reachable Warn if contact is only reachable inside About page content, not directly in footer/nav
8. Do not recommend creating /contact when About or footer already expose contact info

Page-Level Checks (in {{page_checks_html}}), output in this exact order:

Core checks: URL Slug · Title Tag · Meta Description · H1 Tag · Canonical Tag · Image Alt Text · Word Count · Keyword Placement · Heading Structure · Internal Links · Schema (JSON-LD)

Schema (JSON-LD) rules:

• Treat Schema as a quality check, not only a presence check.
• Validate JSON-LD parseability, expected @type, required fields, recommended rich-result fields, nested fields, and primary-type conflicts.
• Fail when JSON-LD is invalid, the expected schema type is missing, required fields are missing, or localized schema clearly points to the wrong language/URL.
• Warn when recommended fields are missing, nested fields are incomplete, multilingual pages lack inLanguage, or localized schema cannot be fully confirmed.
• Pass only when the expected schema type is present, required fields are present, no conflicts are found, and localized schema matches the current page language/URL when applicable.
• For multilingual pages, each language version should have its own schema with matching inLanguage, language-specific headline/description where present, and url / mainEntityOfPage pointing to the current localized canonical URL.

★ Full-only additions:

• OG Tags — og:title, og:description, og:image, og:type, og:url presence and validity
• Twitter Card — twitter:card type, title/description/image (with OG fallback detection)

How to Use Script JSON Output

Same rules across full audit modules — map each field's status directly to the report check table:

• status → pass / warn / fail / error → badge in report
• detail → starting point for Evidence line
• Do not contradict script output unless you have additional observable evidence

For check-site.py output:

• staging_subdomains.status → Staging Subdomain Indexation row status
• staging_subdomains.detail → Staging Subdomain Indexation row detail
• staging_subdomains.public_hosts → evidence for public staging/test hosts
• staging_subdomains.similar_hosts → fail evidence for production-like staging duplicates
• robots.status → robots.txt row status
• sitemap.status → sitemap.xml row status
• sitemap_inventory.status → Sitemap URL Inventory module status
• sitemap_inventory.directories[] → table rows with path, url_count, page_type, and example_url
• sitemap_inventory.detail → short explanatory note below the inventory table

For check-schema.py output:

• status → Schema (JSON-LD) row status
• detail → Schema (JSON-LD) row detail
• parse_errors → fail evidence for malformed JSON-LD
• schemas[].fields_missing → fail evidence for missing required schema fields
• schemas[].recommended_missing and schemas[].nested_issues → warning evidence
• localized_schema.status → language/URL alignment status for multilingual schema
• localized_schema.issues → evidence for schema language or URL mismatch

For check-social.py output:

• og.status → OG Tags row status
• twitter_card.status → Twitter Card row status
• og.fields.* → individual field details for the detail cell
• twitter_card.fields.* → individual field details, note fallback fields

For check-pagespeed.py output:

• status → overall Page Speed status
• category_status → Lighthouse Scores badge status
• metric_status → Core Lab Metrics badge status
• final_url → Final URL line
• categories.performance.score → Performance score
• categories.accessibility.score → Accessibility score
• categories["best-practices"].score → Best Practices score
• categories.seo.score → SEO score
• compact_metrics.fcp.display_value → FCP
• compact_metrics.lcp.display_value → LCP
• compact_metrics.tbt.display_value → TBT
• compact_metrics.cls.display_value → CLS
• compact_metrics.si.display_value → Speed Index
• screenshot truthy → Screenshot available

Do not use overall status for the Lighthouse Scores badge. Category scores use Lighthouse thresholds: 90–100 pass, 50–89 warn, 0–49 fail. Example: Performance 60, Accessibility 84, Best Practices 100, SEO 100 means category_status is warn. If LCP or Speed Index fails, metric_status and overall status may be fail while the Lighthouse Scores panel remains warn.

Inside {{pagespeed_checks_html}}, include a short Priority Actions list after the score/metric cards. Keep it to 2–4 concise items based on failing or warning PageSpeed fields:

• LCP slow → optimize hero media, preload critical image, reduce render-blocking CSS
• Speed Index slow → defer non-critical scripts and reduce above-the-fold JS/CSS
• TBT high → split long tasks and delay third-party tags
• Performance score warning/fail → prioritize the biggest Lighthouse opportunities

Always include the official diagnostic link: https://pagespeed.web.dev/

LLM Review Instructions

Core LLM reviews

Resolve every llm_review_required: true field before writing the report: H1 semantic judgment, Title keyword position, URL Slug evaluation, and Meta Description quality must all receive an explicit judgment.

Full-only LLM checks

OG Tags quality (always review):

og:title   : Does it differ meaningfully from <title>? It should be optimized for social sharing.
og:description : Is it compelling for social feeds? Different focus than meta description is OK.
og:image   : Is the URL an actual image path (not a page URL)?

Twitter Card completeness:

If twitter:card is "summary_large_image", twitter:image (or og:image fallback) must be
at least 300x157px. Flag if the image URL looks like a small icon or favicon.

Recommended Workflow

1. Acknowledge full scope — confirm this is a full audit
2. Infer primary keyword — read the page H1, title, and first paragraph unless the user provided one
3. Phase 1: Run core scripts — check-site → check-page → fetch-page → check-schema
4. Phase 2: Run full-only scripts — verify PageSpeed API key exists, then run check-pagespeed → check-social
5. Core checks — 404 handling, URL canonicalization, E-E-A-T trust pages, i18n/hreflang
6. PageSpeed checks — summarize Lighthouse category scores and lab metrics
7. LLM-only advanced checks — E-E-A-T content quality, duplicate content signals, anchor text quality
8. Summarize findings — Evidence / Impact / Fix format
9. Priority actions — top 5 highest-impact fixes with effort/impact tags
10. Render report — save to reports/<hostname>-<slug>-full-audit.html

Report Detail Writing Rules

Use strict formatting:

Pass → one short phrase. No lists, no elaboration.

Warn → one <div class="detail-issue"> with ≤2 bullet points. One <div class="detail-fix">.

Fail → same as Warn. Lead with the exact failure.

Mandatory Finding Format

**Finding: [Finding Title]**

- **Evidence:** [Observable fact, data point, or marked assumption]
- **Impact:** [SEO / UX consequence]
- **Fix:** [Actionable recommendation with example]

For Priority Actions, add effort/impact tags:

1. [High Impact / Low Effort] Fix og:image — social shares currently show no preview.

Reference Files

• Detailed audit modules and field definitions: references/REFERENCE.md
• Final HTML report template: assets/report-template.html
• PageSpeed validation script: scripts/check-pagespeed.py
• Social tags validation script: scripts/check-social.py
• Core scripts: scripts/ (check-site, check-page, check-schema, fetch-page)