site-geo: An Open-Source GEO Audit Engine for the AI Search Era

site-geo is an open-source audit tool designed specifically for Generative Engine Optimization (GEO). It aims to help overseas sites evaluate their visibility and citeability in AI systems like ChatGPT, Google AI, and Perplexity, addressing the problem that traditional SEO tools cannot meet the core needs of the AI search era.

Traditional SEO tools focus on SERP metrics like keyword rankings and backlink counts, but the rules have fundamentally changed in the AI search environment—AI directly generates answers instead of returning link lists. If content is not understood by AI and included in its knowledge base, even high SEO rankings may lead to missed traffic. The core goal of GEO optimization is to upgrade sites from "searchable" to "citeable", and site-geo's design philosophy is based on this paradigm shift.

site-geo's audit process starts with building a complete site snapshot, systematically analyzing multiple types of pages (such as homepage, about page, service page) and generating profile data including meta-information, content structure, and credibility signals. Its five audit modules include:

1. Visibility Module: Checks AI crawler access configurations (e.g., robots.txt, LLMs.txt);
2. Technical Module: Evaluates infrastructure like HTTPS, SSR, and loading performance;
3. Content Module: Analyzes GEO-friendliness such as information density and E-E-A-T signals;
4. Structured Data Module: Verifies the completeness of JSON-LD markup;
5. Platform Adaptation Module: Evaluates optimization for different AI platforms.

site-geo uses a six-dimensional weighted scoring system to quantify GEO health, with weight distribution as follows: AI Visibility (25%), Brand Authority (20%), Content & E-E-A-T (20%), Technical Foundation (15%), Structured Data (10%), and Platform Adaptation (10%). The featured "AI Cognitive Snapshot" function outputs AI's perception profile of the site (such as the proportion of positive/neutral/controversial perceptions and cognitive keywords) by analyzing content features and entity signals.

site-geo supports full audit mode (page-by-page diagnosis of citeability, content quality, etc.), discovery layer data reuse (to avoid repeated crawling), and AI-enhanced features (semantic analysis for premium members). Deployment is based on Python 3.10+ and FastAPI, with Docker support. It is open-source under the MIT license, with code hosted on GitHub and an online demo available, suitable for overseas brands, cross-border SaaS, etc.

site-geo represents the evolution of search optimization tools toward the GEO era, reminding sites that their optimization goals need to shift from pleasing traditional search engine algorithms to becoming trusted information sources that AI systems are willing to cite. For overseas sites hoping to remain competitive in the global AI search wave, such audit tools will become an important part of their infrastructure.