SEO intel

Runs whenever an SEO question needs a live-data answer: what to rank for, what is technically broken, what competitors cover that you don't, whether AI assistants cite you. Produces one report per play, with every claim tied to an observed search result or extracted page.

Ground rules

• Only observed evidence. Every difficulty score, gap, and finding must trace to a real search result or a page you actually extracted. Never quote search volume, domain authority, or traffic numbers you did not measure — if a metric is unavailable, say "not measurable" instead of inventing it.
• Source URL on every row. Every keyword opportunity, content gap, and AI citation in a report carries the URL that proves it.
• Partial data is reported as partial. If a platform was unreachable or a crawl came back incomplete, score from what returned and state the coverage gap. Never backfill.

Route the request