SEO Data Sources
Use this skill before relying on external SEO data. Read references/data-sources.md (resolved from the suite/plugin root) first.
Workflow
- State the claim that needs proof.
- Pick the smallest source that can prove it.
- Verify whether the provider is installed/authenticated in the current environment.
- If unavailable, mark it as a blocker or hypothesis, not a confirmed finding.
- Attach the data source to each SEO claim.
Provider Roles
- Local scripts: repo/package/site metadata and crawlability baseline.
- Firecrawl: rendered crawl and competitor page extraction.
- Google Search Console: owned-site indexing, queries, impressions, sitemap submission.
- PageSpeed/CrUX: lab and field performance evidence.
- DataForSEO: keyword, SERP, and competitor data via API.
- SE Ranking: packaged SEO/GEO workflows and share-of-voice style reporting.
- Ahrefs: backlinks, content gap, competing pages, and authority estimates.
Output
A short data-source decision record: the claim, the chosen source, why the smaller sources were insufficient, access status (installed / needs auth / unavailable), and cost boundary. Attach it to the audit report or issue.
Failure Handling
- If no available source can prove the claim, report it as
Hypothesiswith the missing provider named as a blocker. Do not downgrade silently to a weaker source without recording the substitution.
Rules
- Do not present optional MCPs as installed without current proof.
- Do not use crawl data as ranking data.
- Do not use Search Console without explicit account/browser/API access.
- Do not put API keys in repo files, reports, prompts, or screenshots.