Verified against Claude · 2026-08-04
Audit a competitor's AI visibility to find your citation gap
Compares your AI-crawler access, llms.txt, and answer-engine citation presence against a named competitor on a shared topic, producing three ranked fixes instead of a vague sense that they are winning.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are a competitive AEO analyst with live browsing/search access. Compare AI visibility between two sites on one shared topic — not a general opinion of "who ranks better." YOUR DOMAIN: example.com COMPETITOR: competitor.com SHARED TOPIC/QUERY: how to calculate customer acquisition cost STEPS 1. Fetch both domains' robots.txt. For each, check whether GPTBot, PerplexityBot, ClaudeBot, and Google-Extended are allowed or blocked, applying real precedence rules (most specific User-agent group wins, longest matching path wins within a group) — not a surface read of the file. Note any difference between the two sites. 2. Check both domains for a public llms.txt, and if one exists, whether its listed pages are actually reachable rather than 404ing or blocked elsewhere in that same domain's robots.txt. 3. Actually search or ask about "how to calculate customer acquisition cost" using your live tool, and report which domain, if either, gets cited, and how — a quoted stat, a linked source, named but not quoted, or not mentioned at all. 4. Based on steps 1-3, identify the single most likely structural reason for any citation gap: crawler access, structured data, missing llms.txt, page freshness, or content specificity — pick the one the evidence actually points to, not the one that's easiest to fix. 5. If both domains show identical crawler access and llms.txt status but one still gets cited and the other doesn't, say so explicitly — that result means the gap is a content or authority difference, not a technical-access one, and the fix recommendations should reflect that instead of repeating a crawler-access fix that wouldn't change anything. OUTPUT 1. A crawler-access table for both domains. 2. The citation-presence result from step 3. 3. Three ranked fixes for example.com, ordered by how much of the gap each is likely to close, given the specific cause identified in step 4.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
This mirrors the AI Visibility Checker's own scoring weights — crawler access 40%, structured data 20%, page basics 20%, llms.txt 10%, sitemap 10% — so a competitor gap almost always traces back to one of exactly five buckets, turning "they show up more" into one concrete, fixable difference instead of a vague impression that's hard to act on. It also acts on the tool's own stated finding that classic Google ranking and AI-crawler access are decided by entirely separate bot rules: a competitor can rank behind you in Google Search while still being the one AI engines cite, because their robots.txt treats GPTBot or PerplexityBot differently than yours does — a gap invisible to any classic rank tracker that only watches Googlebot-driven ranking positions. The explicit branch for identical technical access but different citation outcomes matters because it's the case a purely mechanical audit is most likely to get wrong by default: without that check, a report built around the five-bucket technical framing will keep recommending technical fixes even when the evidence in front of it already shows the technical layer is a dead end and the real gap is content depth or topical authority — naming that honestly is what keeps the ranked fixes from wasting effort on the wrong layer.
What you get back
| | example.com | competitor.com | |---|---|---| | GPTBot | Allowed | Allowed | | PerplexityBot | Blocked | Allowed | | ClaudeBot | Allowed | Allowed | | Google-Extended | Allowed | Allowed | | /llms.txt | Not found | Found, and its 3 listed pages are all live and reachable | Citation result: competitor.com was cited by Perplexity with a quoted definition; example.com was not mentioned. Ranked fixes: 1) unblock PerplexityBot — this alone likely explains most of the gap on that specific engine, 2) publish /llms.txt pointing to your top 3 pages, 3) add a definition-style opening sentence matching the cited competitor pattern.
Verified against
Claude Claude Sonnet 4.5 · 2026-08-04
Perplexity Sonar Pro · 2026-08-04
Changelog
- 2026-08-04 — Initial publish. Anchored the five-bucket framing to the AI Visibility Checker's own scoring weights so the audit produces the same vocabulary the tool's report uses.
- 2026-08-06 — Added the identical-access-but-different-outcome branch after a test case kept recommending crawler-access fixes even though both domains already showed the same technical access.
Building this for real?
This is a free starting point. If you'd rather have SEO built and running for your business, that's Scult's day job.
EXPLORE SEO
