Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Master orchestrator for a full SEO audit suite powered by the Ahrefs MCP. Use this skill when running a comprehensive SEO audit, scoping a quarterly health check, doing pre-acquisition SEO due diligence, or post-migration verification. Triggers on full SEO audit, comprehensive SEO review, SEO health check, audit my site, SEO due diligence, audit suite, comprehensive audit, end-to-end SEO. Also triggers when a stakeholder wants the complete picture rather than a single-dimension audit.
.claude/skills/rampstackco-seo-audit-orchestration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 80% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 73% | 0% |
Run a complete SEO audit by sequencing the sibling audit skills in a defined order. Stack-agnostic. Assumes the Ahrefs MCP is connected. Produces a single rollup report that synthesizes findings across backlinks, keywords, content, traffic, technical health, and rankings.
seo-onpage, seo-backlink-audit, etc.)seo-traffic-diagnosis)seo-keyword-gap-audit)seo-rank-tracking)A complete audit moves through six phases in order. Skipping phases produces gaps. Reordering them produces rework.
Define the audit before running it.
Output: a 1-page audit charter.
Pull the raw data from Ahrefs and any companion sources.
Required pulls:
Companion pulls (not Ahrefs-native):
Document data freshness for every pull. Stale data yields wrong conclusions.
Run each child audit. Each produces its own findings doc.
| Sub-audit | Skill | What it produces | | --- | --- | --- | | Site health | seo-site-health-audit | Prioritized technical fix backlog | | Backlinks | seo-backlink-audit | Profile health, toxic list, reclamation list | | Keywords | seo-keyword-gap-audit | Prioritized opportunity list | | Content | seo-content-gap-audit | Create/update/merge roadmap | | Page-level | seo-onpage | Per-page audit on top pages | | AI search | seo-aeo-geo | AI search readiness gaps |
The child skills do the analysis. This skill sequences them and integrates the outputs.
Combine findings into themes. A list of 200 issues is not an audit. A short list of themes is.
Themes typically emerge in categories like:
Each theme should answer: what is happening, why it matters, what is the size of prize, what is the fix.
Rank the themes. Use a simple impact/effort matrix.
Tie each theme to a measurable target (organic clicks, ranked keywords in target band, conversions from organic).
Produce the rollup report. See references/audit-rollup-template.md.
Structure:
Walk stakeholders through it. Get commitment to the next 90 days of work.
A rollup audit report with:
Total length, executive summary length, and theme count all vary by audit type: a quarterly health check is not a due-diligence audit. The length-targets table in references/audit-rollup-template.md owns those numbers per type. Whatever the type, the executive summary stays readable in 5 minutes.
This skill's output depends on data, measurements, or tool results it cannot generate on its own. When a required input, tool, or data source is unavailable or unverifiable, the sanctioned output is the deliverable with the gap stated: what was needed, what was actually obtained or verified, and which parts of the output are affected. Fabricating, estimating, or interpolating a required number to complete the deliverable is never sanctioned. A stated gap is a complete answer.
references/audit-rollup-template.md - Template for the rollup report including executive summary, theme structure, and roadmap format.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 10,002 | 4,482 | -55% | 1 | 1 | 0% | 1,386 | 2,622 | +89% | 0 | 0 | — |
case-01 | fail→pass | 34,438 | 36,162 | +5% | 1 | 1 | 0% | 6,255 | 8,133 | +30% | 0 | 0 | — |
case-02 | fail→fail | 34,986 | 34,902 | -0% | 1 | 1 | 0% | 6,260 | 8,139 | +30% | 0 | 0 | — |
case-03 | fail→fail | 37,152 | 35,342 | -5% | 1 | 1 | 0% | 6,261 | 7,419 | +18% | 0 | 0 | — |
case-04 | fail→fail | 38,354 | 26,284 | -31% | 1 | 1 | 0% | 6,203 | 6,326 | +2% | 0 | 0 | — |
case-05 | fail→fail | 23,032 | 21,674 | -6% | 1 | 1 | 0% | 4,515 | 5,672 | +26% | 0 | 0 | — |
case-06 | fail→pass | 27,664 | 14,451 | -48% | 1 | 1 | 0% | 6,186 | 4,432 | -28% | 0 | 0 | — |
case-07 | pass→pass | 11,731 | 7,461 | -36% | 1 | 1 | 0% | 1,894 | 3,180 | +68% | 0 | 0 | — |
case-08 | pass→pass | 11,833 | 11,888 | +0% | 1 | 1 | 0% | 1,973 | 3,774 | +91% | 0 | 0 | — |
case-09 | fail→pass | 8,360 | 3,438 | -59% | 1 | 1 | 0% | 1,440 | 2,403 | +67% | 0 | 0 | — |
case-10 | pass→pass | 5,601 | 3,854 | -31% | 1 | 1 | 0% | 968 | 2,533 | +162% | 0 | 0 | — |
case-11 | pass→pass | 7,836 | 4,001 | -49% | 1 | 1 | 0% | 1,343 | 2,529 | +88% | 0 | 0 | — |
case-12 | pass→pass | 11,120 | 7,109 | -36% | 1 | 1 | 0% | 1,730 | 3,088 | +78% | 0 | 0 | — |
case-13 | fail→pass | 8,004 | 2,874 | -64% | 1 | 1 | 0% | 1,287 | 2,316 | +80% | 0 | 0 | — |
case-14 | fail→pass | 7,754 | 2,102 | -73% | 1 | 1 | 0% | 1,305 | 2,262 | +73% | 0 | 0 | — |
case-20 | fail→pass | 11,410 | 6,729 | -41% | 1 | 1 | 0% | 1,830 | 3,012 | +65% | 0 | 0 | — |
case-15 | fail→pass | 4,442 | 2,897 | -35% | 1 | 1 | 0% | 725 | 2,346 | +224% | 0 | 0 | — |
case-16 | pass→pass | 17,721 | 16,023 | -10% | 1 | 1 | 0% | 2,734 | 4,356 | +59% | 0 | 0 | — |
case-17 | pass→pass | 9,831 | 6,994 | -29% | 1 | 1 | 0% | 1,622 | 3,075 | +90% | 0 | 0 | — |
case-18 | fail→pass | 9,563 | 5,525 | -42% | 1 | 1 | 0% | 1,430 | 2,782 | +95% | 0 | 0 | — |
case-19 | pass→pass | 11,898 | 7,492 | -37% | 1 | 1 | 0% | 2,014 | 3,189 | +58% | 0 | 0 | — |
case-22 | pass→pass | 17,158 | 12,773 | -26% | 1 | 1 | 0% | 2,344 | 3,794 | +62% | 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 +36 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.