The Most Starred Open-Source Generative Engine Optimization (GEO) Tools in 2026: AI Visibility Trackers, Skills and Self-Hosted Options

The most starred open-source generative engine optimization (GEO) tools on GitHub are open-seo-mcp-skills (9,060 stars), GEOFlow (4,056) and aaron-marketing-skills (3,115), while elmo leads tools built purely for AI visibility tracking with 584 stars.

Open-source GEO tools do different jobs: some track mentions and citations, some audit and fix websites, and others provide agent skills or content infrastructure. GitHub stars measure attention, not quality or use. The ranking below separates those functions and identifies hosted dependencies.

What are the most starred open-source generative engine optimization tools?

Which open-source GEO tools track citations, fix websites or support content work?

The twelve ranked tools fall into four groups, and they answer different questions:

• Visibility trackers — elmo, limelit open, GetCito, geolook: is my brand mentioned or cited when people ask AI assistants about my category? They run a fixed set of prompts on a schedule and count.
• Site and content fixers — geo-optimizer-skill, GEORank, geo-doctor: can AI engines read and cite my pages, and what should I change?
• Agent skill kits — open-seo-mcp-skills, aaron-marketing-skills, gtm-engineer-skills: workflows for Claude Code or Codex where GEO sits beside SEO and content work.
• Infrastructure — GEOFlow for producing and publishing content, GEOInflow for seeing what AI crawlers fetch.

If the question is are we being cited, start with a tracker. If it is why not, start with a fixer. A star count does not distinguish those jobs, so the most starred repository is not necessarily the closest match for your task.

What does each ranked open-source GEO tool do?

1. open-seo-mcp-skills — 9,060 stars

Eight MIT-licensed skills for Claude covering keyword research, rank tracking, audits, backlinks and AI visibility, run on your own Search Console, GA4, Google Ads and DataForSEO data. Its ai-visibility skill reports AI-referral traffic from GA4. The skills are open; the connector they run on is Ryze's hosted MCP server, so it is not self-contained. Most of it is SEO; GEO is one skill of eight.

2. GEOFlow — 4,056 stars

GEOFlow is a self-hosted Laravel system (Docker Compose, Postgres, Redis) for producing and distributing content at scale: knowledge import, AI article generation, AI quality checks, human review, publishing to several sites, automatic robots.txt, sitemap and llms.txt, and AI visibility analysis. Its README is in Chinese. Its licence is AGPL-3.0, with a commercial licence offered separately.

3. aaron-marketing-skills — 3,115 stars

aaron-marketing-skills provides 120 Markdown skills across seven marketing disciplines, installable as a Claude Code plugin or with npx skills add. GEO is a part of it: a 16-skill SEO and GEO track that includes a GEO content optimiser and a content-quality auditor. The README says outcomes on real projects are not yet validated.

4. gtm-engineer-skills — 1,359 stars

gtm-engineer-skills provides twelve skills for Claude Code and Codex that run as a pipeline: brand and prompt research, content planning and drafting, charts an AI can read, claim audits, and an AEO site audit that writes code fixes. The "16 foundational checks" in its description belong to the audit skill.

5. geo-optimizer-skill — 1,302 stars

geo-optimizer-skill is a Python CLI, MCP server and Claude Code skill that scores a site from 0 to 100 on whether ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews can crawl and cite it, then writes the fixes: robots.txt, llms.txt, JSON-LD and meta tags. geo citations asks the real engines whether your brand is cited, with your own API keys. It is the open engine behind a paid hosted product, GeoReady.

6. GEOInflow — 829 stars

GEOInflow provides self-hosted analytics for the AI side of your traffic: which AI crawlers fetch which pages, failed requests, crawl coverage against your sitemap, and visits and conversions referred by AI assistants. Its README is careful to say that crawl logs alone do not prove citations or rankings. Created on 26 October 2026, it is the youngest tool on the list.

7. geolook — 812 stars

geolook is a Python dashboard you run on one machine. It samples 17 engines — ten through APIs, seven by hand — and measures mention rate, rank and citation share, then audits the site and turns gaps into tickets, assets and published fixes it re-checks. It covers Chinese engines in depth.

8. geo-doctor — 601 stars

geo-doctor is a bilingual Chinese and English agent skill with an eight-step workflow: site check, fact card, question set, sampling answers, competitor gap diagnosis, batch writing, publishing and a re-test. It covers Doubao, DeepSeek, Yuanbao, Qwen, Wenxin and Kimi as well as ChatGPT, Claude, Perplexity, Gemini and Copilot, and runs through npx.

9. elmo — 584 stars

elmo is the most starred tool built purely to track AI visibility. It runs your prompts several times a day on ChatGPT, Google AI Mode and AI Overviews, Gemini, Perplexity and Copilot, and on Claude, Grok and others through their APIs, and tracks visibility, share of voice, sentiment, citations and query fan-out. A paid Elmo Cloud is offered alongside.

10. GEORank — 534 stars

Despite the name, GEORank is a diagnosis and planning workbench rather than a rank tracker: it checks a site's schema, meta tags, structure and citation signals, and generates Q&A content, 30/60/90-day plans, keyword expansions, JSON-LD and llms.txt. Its README states that it does not sell rankings.

11. limelit open — 438 stars

limelit open is a single Go binary with SQLite that tracks brand mentions and citations in ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews, AI Mode and Bing Copilot, reporting visibility, share of voice, position and citation share, with 14 MCP tools. It is the open core of Limelit Cloud, and its README marks it as pre-release.

12. GetCito — 426 stars

GetCito is a self-hosted tracker for prompts, mentions, sentiment, citations and competitors across the answer engines you configure, with growth tasks, content audits, an API, an MCP server and a Looker connector.

How were the open-source GEO tools selected and ranked?

Two GitHub searches and one filter, nothing added by hand. The population is every repository that either has the exact phrase "generative engine optimization" in its name or description, or carries the topic tag generative-engine-optimization: 1,525 repositories. Of those, 1,025 name GEO or generative engine optimization in their description. Curated lists and guides (102) and archived repositories (8) were removed, leaving 915 tools, ranked by stars. The twelve above are every one with more than 400.

Both conditions matter. The phrase search alone misses GEOFlow, geo-optimizer-skill and limelit, which call themselves GEO tools and carry the tag but never spell out the full phrase. The tag alone pulls in general marketing kits that tag themselves for reach; the description filter keeps out the ones that never mention GEO.

The population is mostly new, with stars concentrated in a small group. Of the 915 GitHub repositories that describe themselves as GEO tools, 858 were created in 2026. 530 have no stars at all, and only 25 have 100 or more. The twelve above hold 84% of all the stars the 915 have received. Those figures describe repository age and attention, not demonstrated adoption.

Why are the original GEO paper's code and AutoGEO outside the ranking?

GEO-optim/GEO, the code and GEO-bench data behind the 2023 paper that named the field, has 350 stars and would rank 13th. Its repository has no description and no topic tags, so neither search finds it. It is research code for reproducing the paper rather than a tool you point at your own site. AutoGEO (237 stars), the ICLR 2026 paper's code that learns which rewrites engines prefer, is in the population but falls below the cut.

What can this ranking tell you—and what can it not?

• Stars measure attention, not quality or use. Several of these repositories are weeks old; GEOInflow was created 15 days before this count.
• Self-description defines the field here. A tool that does the same job without calling itself GEO, or without the tag, is not counted.
• Open source varies. open-seo-mcp-skills depends on a hosted connector, and limelit open, geo-optimizer-skill and elmo are open cores of paid products.
• Claims are the authors' own. Benchmarks and lift figures quoted in these READMEs were not reproduced for this list.

Where can you verify the repositories and star counts?

Every star count is the repository's stargazers_count as returned by the GitHub API on 10 November 2026; it will have moved since.

• Ryze-AI-Adgent/open-seo-mcp-skills
• yaojingang/GEOFlow
• aaron-he-zhu/aaron-marketing-skills
• onvoyage-ai/gtm-engineer-skills
• Auriti-Labs/geo-optimizer-skill
• yaojingang/GEOInflow
• aigclink/geolook
• Fangx-AI/geo-doctor
• elmohq/elmo
• yaojingang/GEORank
• limelit-co/open
• ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tool

How do GitHub stars differ from AI search citations?

A star is a developer bookmarking a repository. Being cited by an AI search engine is what most of these tools measure: whether an assistant names or links your page when someone asks a question. For what that measurement involves, see what AI visibility is and how to track brand mentions in AI search.

FAQ

Frequently asked questions

What is the most starred open-source GEO tool?

open-seo-mcp-skills by Ryze AI, with 9,060 GitHub stars on 10 November 2026, followed by GEOFlow (4,056) and aaron-marketing-skills (3,115). Among tools built only to track AI visibility, elmo leads with 584.

Is there a self-hosted generative engine optimization tool?

Several. elmo, limelit open, GetCito and geolook track AI visibility on your own server; GEOFlow produces and publishes content; GEOInflow analyses AI crawler traffic; GEORank diagnoses site readiness.

Which open-source tool tracks whether ChatGPT cites my brand?

elmo, limelit open, GetCito and geolook all run prompts against ChatGPT and other assistants on a schedule and record mentions and citations; geo-optimizer-skill can also ask the engines directly with your API keys.

How many GEO tools are on GitHub?

915 repositories describe themselves as GEO tools as of 10 November 2026, excluding curated lists and archived repositories. 858 were created in 2026, 530 have no stars, and 25 have 100 or more.

Where is the code for the original GEO paper?

GEO-optim/GEO on GitHub, with 350 stars. It is not ranked here because its repository has no description or topic tags.

What is the short answer for choosing an open-source GEO tool?

On 10 November 2026 the most starred open-source GEO tools on GitHub were open-seo-mcp-skills (9,060 stars), GEOFlow (4,056) and aaron-marketing-skills (3,115); the most starred dedicated AI visibility tracker was elmo (584). Of 915 repositories that call themselves GEO tools, 858 are from 2026 and 530 have no stars, so the list will move quickly.