Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Technical SEO auditing covering crawlability, indexation, Core Web Vitals, on-page optimization, and competitive gaps, with an 85-point checklist and remediation plans. Use when auditing technical SEO or diagnosing indexation issues.
.claude/skills/borghei-seo-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 47% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 80% | 0% |
Production-grade SEO audit framework with an 85-point checklist across 8 dimensions, severity-weighted scoring, automated diagnostic workflows, and prioritized remediation plans. Covers technical SEO, on-page optimization, content quality, competitive positioning, and migration readiness.
Pick the operating mode that matches the situation:
Before auditing, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
bash# Check redirect chains and status codes python scripts/redirect_checker.py --url https://example.com/old-page --json # Analyze XML sitemap for errors python scripts/sitemap_analyzer.py --sitemap https://example.com/sitemap.xml # Score content quality for SEO python scripts/content_scorer.py article.md --json
Load the reference that matches the task — keep this file lean and pull detail on demand:
In scope: technical SEO across all 8 audit dimensions, severity-weighted scoring with prioritized remediation, traffic-drop diagnosis, competitive gap analysis, pre/post-migration checklists, AI content quality detection.
Out of scope: content creation or rewriting (use Content Creator), structured data implementation (use Schema Markup), site architecture redesign (use Site Architecture), link building execution, paid search audits, server/CDN provisioning.
Known limitations: field CWV data requires sufficient traffic for CrUX; off-page signals need third-party tools (Ahrefs, SEMrush); AI Overview impact on CTR varies by query type; Google's algorithm changes 500–600 times/year, so findings are point-in-time.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 21,018 | 10,917 | -48% | 1 | 1 | 0% | 2,883 | 2,729 | -5% | 0 | 0 | — |
case-02 | fail→fail | 12,554 | 15,669 | +25% | 1 | 1 | 0% | 2,984 | 4,070 | +36% | 0 | 0 | — |
case-11 | fail→pass | 13,103 | 2,482 | -81% | 1 | 1 | 0% | 2,355 | 1,452 | -38% | 0 | 0 | — |
case-03 | fail→pass | 15,092 | 15,368 | +2% | 1 | 1 | 0% | 2,234 | 3,284 | +47% | 0 | 0 | — |
case-04 | fail→pass | 12,374 | 7,355 | -41% | 1 | 1 | 0% | 1,974 | 2,240 | +13% | 0 | 0 | — |
case-05 | fail→fail | 20,045 | 30,020 | +50% | 1 | 1 | 0% | 3,073 | 5,744 | +87% | 0 | 0 | — |
case-06 | fail→fail | 26,553 | 36,108 | +36% | 1 | 1 | 0% | 4,038 | 7,325 | +81% | 0 | 0 | — |
case-12 | fail→pass | 7,068 | 3,196 | -55% | 1 | 1 | 0% | 901 | 1,622 | +80% | 0 | 0 | — |
case-07 | fail→fail | 21,497 | 35,331 | +64% | 1 | 1 | 0% | 3,642 | 7,319 | +101% | 0 | 0 | — |
case-08 | fail→pass | 12,325 | 17,650 | +43% | 1 | 1 | 0% | 1,876 | 3,947 | +110% | 0 | 0 | — |
case-09 | fail→pass | 11,721 | 1,925 | -84% | 1 | 1 | 0% | 1,914 | 1,377 | -28% | 0 | 0 | — |
case-10 | fail→pass | 5,157 | 2,015 | -61% | 1 | 1 | 0% | 692 | 1,399 | +102% | 0 | 0 | — |
case-13 | fail→pass | 10,010 | 3,057 | -69% | 1 | 1 | 0% | 1,556 | 1,587 | +2% | 0 | 0 | — |
case-14 | fail→pass | 10,860 | 10,556 | -3% | 1 | 1 | 0% | 1,937 | 2,705 | +40% | 0 | 0 | — |
case-15 | fail→pass | 12,682 | 9,511 | -25% | 1 | 1 | 0% | 1,979 | 2,501 | +26% | 0 | 0 | — |
case-16 | pass→pass | 6,272 | 3,686 | -41% | 1 | 1 | 0% | 842 | 1,741 | +107% | 0 | 0 | — |
case-17 | pass→pass | 10,734 | 6,937 | -35% | 1 | 1 | 0% | 1,787 | 2,087 | +17% | 0 | 0 | — |
case-18 | pass→pass | 7,037 | 2,042 | -71% | 1 | 1 | 0% | 1,164 | 1,453 | +25% | 0 | 0 | — |
case-19 | fail→pass | 3,370 | 1,590 | -53% | 1 | 1 | 0% | 415 | 1,301 | +213% | 0 | 0 | — |
case-20 | fail→pass | 12,085 | 3,525 | -71% | 1 | 1 | 0% | 1,656 | 1,658 | +0% | 0 | 0 | — |
case-21 | pass→pass | 9,282 | 3,616 | -61% | 1 | 1 | 0% | 1,401 | 1,561 | +11% | 0 | 0 | — |
case-22 | pass→pass | 11,189 | 2,509 | -78% | 1 | 1 | 0% | 1,682 | 1,461 | -13% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.