← ClaudeAtlas

geo-citation-trackerlisted

Use when the user wants to measure or monitor whether AI answer engines actually name and cite their brand over time — not whether a page is optimized, but what the engines are saying right now. Also use when the user mentions AI share of voice, LLM citation tracking, brand visibility in ChatGPT/Perplexity/Claude/AI Overviews, "are we showing up in AI answers", answer-engine monitoring, or wants to prove GEO work moved the needle. Designs the prompt set, scores the run, and says whether a change between runs is signal or noise.
sarojkjha/aaj-marketing-skills · ★ 0 · AI & Automation · score 73
Install: claude install-skill sarojkjha/aaj-marketing-skills
# GEO Citation Tracker Measure what AI engines actually say about a brand, on a fixed prompt set, over time. `seo-geo-aeo-audit` scores whether a page is *built* to be cited. This skill measures whether it *is* — and whether the number is moving. The two are routinely confused, and a brand can score an A on readiness while appearing in one answer out of forty. **The distinction this skill exists to enforce is presence versus citation.** Being *named* in an answer means the engine knows the brand from its training priors. Being *cited* means the engine read a specific page and linked it as a source. Only the second is something content work can reliably influence, and only the second compounds. Most brands track the first, report it as GEO performance, and can't explain why publishing more changes nothing. **The failure mode this skill exists to prevent is reading noise as movement.** AI answers are non-deterministic — the same prompt returns different brands on different days. On a 20-prompt set, presence moving from 30% to 40% is entirely consistent with nothing having happened at all. That delta gets put in a board deck. The engine here refuses to let it: it computes what size of change the prompt set could actually detect, and says plainly when the movement is inside the noise. ## When to use The user wants to know their current standing in AI answers, wants to set up ongoing monitoring, or wants to demonstrate that GEO work produced a result. If they want to know wh