Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Diagnose and audit SEO issues affecting crawlability, indexation, rankings, and organic performance. Use when the user asks for an SEO audit, technical SEO review, ranking diagnosis, on-page SEO review, meta tag audit, or SEO health check. This skill identifies issues and prioritizes actions but does not execute changes. For large-scale page creation, use programmatic-seo. For structured data, use schema-markup.
.claude/skills/dokhacgiakhoa-seo-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 28% | 0% |
You are an SEO diagnostic specialist. Your role is to identify, explain, and prioritize SEO issues that affect organic visibility—not to implement fixes unless explicitly requested.
Your output must be evidence-based, scoped, and actionable.
Before performing a full audit, clarify:
If critical context is missing, state assumptions explicitly before proceeding.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 26,411 | 37,209 | +41% | 1 | 1 | 0% | 4,042 | 5,214 | +29% | 0 | 0 | — |
case-02 | fail→pass | 41,422 | 22,295 | -46% | 1 | 1 | 0% | 6,673 | 4,531 | -32% | 0 | 0 | — |
case-03 | fail→pass | 39,959 | 27,831 | -30% | 1 | 1 | 0% | 6,271 | 4,878 | -22% | 0 | 0 | — |
case-04 | pass→pass | 7,363 | 4,407 | -40% | 1 | 1 | 0% | 1,269 | 1,747 | +38% | 0 | 0 | — |
case-05 | pass→pass | 12,428 | 9,826 | -21% | 1 | 1 | 0% | 1,774 | 2,446 | +38% | 0 | 0 | — |
case-06 | pass→pass | 14,974 | 15,583 | +4% | 1 | 1 | 0% | 2,199 | 3,330 | +51% | 0 | 0 | — |
case-07 | pass→pass | 15,445 | 12,521 | -19% | 1 | 1 | 0% | 2,375 | 2,997 | +26% | 0 | 0 | — |
case-08 | pass→pass | 16,408 | 18,982 | +16% | 1 | 1 | 0% | 2,507 | 3,803 | +52% | 0 | 0 | — |
case-09 | pass→pass | 12,002 | 5,092 | -58% | 1 | 1 | 0% | 1,961 | 1,643 | -16% | 0 | 0 | — |
case-10 | pass→fail | 14,865 | 13,377 | -10% | 1 | 1 | 0% | 2,356 | 3,022 | +28% | 0 | 0 | — |
case-11 | pass→pass | 12,968 | 2,186 | -83% | 1 | 1 | 0% | 2,168 | 1,303 | -40% | 0 | 0 | — |
case-12 | fail→pass | 14,465 | 16,550 | +14% | 1 | 1 | 0% | 2,183 | 3,578 | +64% | 0 | 0 | — |
case-13 | pass→pass | 13,577 | 2,320 | -83% | 1 | 1 | 0% | 2,299 | 1,284 | -44% | 0 | 0 | — |
case-14 | pass→pass | 11,760 | 8,539 | -27% | 1 | 1 | 0% | 1,855 | 2,359 | +27% | 0 | 0 | — |
case-15 | pass→pass | 9,010 | 2,068 | -77% | 1 | 1 | 0% | 1,392 | 1,204 | -14% | 0 | 0 | — |
case-16 | pass→pass | 12,923 | 7,603 | -41% | 1 | 1 | 0% | 1,813 | 2,126 | +17% | 0 | 0 | — |
case-17 | pass→pass | 16,778 | 8,909 | -47% | 1 | 1 | 0% | 2,400 | 2,251 | -6% | 0 | 0 | — |
case-18 | pass→pass | 23,652 | 22,114 | -7% | 1 | 1 | 0% | 3,230 | 3,873 | +20% | 0 | 0 | — |
case-19 | fail→fail | 18,258 | 16,147 | -12% | 1 | 1 | 0% | 2,530 | 3,693 | +46% | 0 | 0 | — |
case-20 | fail→pass | 9,518 | 6,748 | -29% | 1 | 1 | 0% | 1,610 | 2,004 | +24% | 0 | 0 | — |
case-21 | fail→fail | 12,778 | 14,050 | +10% | 1 | 1 | 0% | 2,435 | 3,383 | +39% | 0 | 0 | — |
case-22 | fail→fail | 11,466 | 18,109 | +58% | 1 | 1 | 0% | 2,029 | 3,898 | +92% | 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 +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.