Short answer: EchoRank is a GEO and SEO platform you can self-host. It measures how a website appears in AI answer engines such as ChatGPT, Google Gemini, Anthropic Claude and Perplexity, produces a GEO (Generative Engine Optimization) and SEO score, and ranks the fixes by impact.
What is EchoRank?
EchoRank is a GEO and SEO platform I built. It runs bounded, cached query checks against ChatGPT, Gemini, Claude and Perplexity through their search-enabled APIs, and records citations, mentions, positions and competing sources. Those model checks need your own provider API keys.
GEO vs. SEO: what is the difference?
SEO aims to rank your pages in search results and earn clicks. GEO aims to get your content cited and mentioned in AI-generated answers. The two overlap: crawlable pages, clear structure and trustworthy sources help both. A 2023 paper, “GEO: Generative Engine Optimization” by Aggarwal et al., found that adding citations, quotations and statistics improved how sources were surfaced in generative engine responses.
| Topic | SEO | GEO |
|---|---|---|
| Goal | Rank in search results and earn clicks | Be cited and mentioned in AI-generated answers |
| Key signals | Keywords, links, technical health | Direct-answer blocks, sources, structured data, bot access |
| Crawlers that matter | Googlebot, Bingbot | GPTBot, ClaudeBot, PerplexityBot and other AI crawlers |
| How to measure | Rankings, impressions, clicks | Query-by-model checks, citation and mention tracking |
What EchoRank scores
The GEO score runs from 0 to 100 and rests on four pillars:
- AI crawler access: robots.txt rules for each AI bot (RFC 9309), WAF and CDN blocks, and whether pages render on the server or only in the browser.
- LLM readability: direct-answer and FAQ blocks, information density, Schema.org coverage and llms.txt.
- Trust and E-E-A-T: author information, HTTPS and HSTS, contact and policy pages, and factual consistency.
- Technical SEO: titles, meta tags, canonical URLs, broken links, thin content and internal links.
Key features
- Multi-model visibility tracking: query-by-model checks that record citations, mentions, positions and competing sources.
- AutoGEO Content Studio: scores content on six metrics (statistics, quotations, citations, answer capsules, authority and fluency, and keyword stuffing) and proposes block-level rewrites as diffs.
- AI bot simulator: fetches your pages as GPTBot, ClaudeBot, PerplexityBot and others, detects WAF challenges and empty client-side shells, and generates allow rules for Cloudflare, nginx, Apache and AWS WAF.
- Knowledge graph: extracts products, brands, categories, FAQs, authors and organizations, reconciles them with Wikidata and outputs Schema.org JSON-LD @graph.
- Competitor intelligence: Share of Model against the competitors you configure.
- AI distribution: IndexNow, sitemap resubmission through Google Search Console, llms.txt, llms-full.txt, and RSS, Atom and JSON feeds.
- Scheduled reports: daily or weekly HTML, PDF or JSON reports delivered by email or webhook.
How to run it locally
With Node.js 22 or newer, you can start from your own machine:
git clone https://github.com/kobeumut/EchoRank.git
cd EchoRank
npm ci
cp .env.example .env
npm run db:migrate
npm run devOpen http://localhost:3000, add a site and click Add and audit. Then open What to Do? first: it lists prioritized fixes with evidence and ready-made JSON-LD, robots.txt and llms.txt snippets. Python 3.10 or newer is optional but recommended for the crawler and scoring engine.
English and Turkish interface
The interface works in English (the default) and Turkish. Switch languages with the EN / TR control in the header. Audit text is written in the language of the request that starts the audit.
License and source code
The source is at github.com/kobeumut/EchoRank, licensed under CC BY-NC-SA 4.0. You may share and adapt it with attribution, and adaptations must use the same license. Commercial use is not allowed.
Frequently asked questions
What does EchoRank measure?
It crawls your site, scores GEO and SEO on a 0 to 100 scale, checks whether AI bots can reach your pages, and tracks whether ChatGPT, Gemini, Claude or Perplexity cite your site for the queries you track.
Where should I start with GEO?
Confirm that AI crawlers can reach your pages (robots.txt and WAF rules) and that your main content is in the server-rendered HTML. Then add a short direct-answer block, sourced statistics and clear author information to your key pages.
What is llms.txt?
llms.txt is a Markdown file that summarizes a site’s key sections and pages for language models, so they can understand the site more easily. EchoRank generates llms.txt and llms-full.txt from your crawl data.
Can I run it on my own server?
Yes. SQLite is used for development and PostgreSQL for production. The Python crawler and scoring engine is a separate package; it is optional but recommended.
Can I use it commercially?
No. The CC BY-NC-SA 4.0 license prohibits commercial use. Check what the license allows and contact the licensor if you need more.
Author: Umut ADALI, iOS and Android developer and the creator of EchoRank. To contribute, read CONTRIBUTING.md in the repository.