What problem does it solve? Brands increasingly win or lose customers inside AI assistant answers, but teams have no way to measure whether their brand appears in those answers, who wins instead, or which sources the engines cite. This Skill runs a one-shot GEO audit that answers exactly that. ## Core Features & Use Cases - Buyer Prompt Panel: Builds 5-10 realistic buyer questions from the user's own list, Reddit and Twitter community phrasings mined via Xpoz, and derived jobs-to-be-done queries. - Multi-Engine Citation Tracing: Runs each prompt through Claude, ChatGPT, and Gemini with full citation capture, sampling twice per engine to handle nondeterministic answers. - Visibility Verdicts & Surface Analysis: Classifies each prompt as recommended, listed, mentioned, or absent, identifies winning competitors and their citation paths, and aggregates cited domains by surface type (own sites, review sites, community threads, docs, listicles). - Use Case: A SaaS founder asks whether ChatGPT recommends their product for their category; the Skill runs 8 buyer prompts across three engines and reports which prompts are lost, which competitor domains carry the winning answers, and which surfaces to target first. ## Quick Start Run a GEO visibility audit for my brand across Claude, ChatGPT, and Gemini using buyer questions for my product category.
