Best free and open-source AI visibility and AEO tools (2026)
Echo is the longest-running open-source AI visibility and answer engine optimization platform: MIT-licensed, free to self-host, and auditable down to how each metric is computed, with 36 releases since July 2025. The field grew fast in 2026: Ansvisor has drawn the most contributors, and Canonry, Limelit, and OneGlanse each take a different approach. Licenses differ more than features do, so check them first.
Key takeaways
- Echo is the longest-running open-source answer engine optimization platform, MIT-licensed and free to self-host, with 36 releases since July 2025 and coverage of ChatGPT, Google AI Mode, Google AI Overviews, Gemini, Perplexity, Copilot, Claude, and any OpenRouter model.
- On public GitHub data as of 17 October 2026, Echo leads this list on stars, releases, and length of continuous history. Ansvisor has the most contributors, and Canonry, Ansvisor, and Echo each logged hundreds of commits in the last 90 days.
- Licenses differ more than features. Echo, Ansvisor, Canonry, and OneGlanse are MIT; Limelit is Apache-2.0; Cited is AGPL-3.0 and Gego is GPL-3.0, which add conditions on modified versions.
- Open source protects you from shutdowns, but not from a project going closed: Notra took its AGPL repository private in October 2026. Prefer permissive licenses and more than one active contributor.
- GetCito is a fork of Echo since July 2026 that has fallen behind it, so we leave it off this list.
- "Free to self-host" is not free to run. You supply the LLM API keys and the infrastructure.
Why open source matters for answer engine optimization
Most AI visibility tools are closed and hosted: you send them your prompts and trust the score they hand back. With open source you can read how a metric is built, and you can run the whole thing on your own infrastructure, so your prompts and history never leave your environment.
For a number that might land in a board report or shape a content budget, being able to audit it matters. So does owning your data outright, with no vendor holding your visibility history and nothing to migrate off if you decide to leave. The tradeoff is upkeep: you run the infrastructure and keep it current as engines change.
The best open-source AEO tools at a glance
| Tool | License | Managed cloud | Best for |
|---|---|---|---|
| Echo | MIT | Yes, from $29/mo | Most teams: every essential AEO feature, the best UX and agent experience |
| OneGlanse | MIT | No | A fully free setup, if you can keep its scrapers running |
| GEO/AEO Tracker | MIT | No | A quick check from a single browser |
| Gego | GPL-3.0 | Waitlist | Bare-bones scheduled checks from a CLI |
By the numbers
| Project | Stars | Commits | Contributors | Beyond the lead dev | Releases | First commit | Commits, last 90d |
|---|---|---|---|---|---|---|---|
| Echo | 462 | 1,158 | 14 | 13 | 36 | Jul 2025 | 374 |
| GEO/AEO Tracker | 295 | 32 | 1 | 0 | 9 | Feb 2026 | 10 |
| OneGlanse | 220 | 523 | 2 | 1 | 0 | Apr 2026 | 27 |
| Limelit | 192 | 79 | 2 | 1 | 13 | Sep 2026 | 79 |
| Canonry | 186 | 1,135 | 8 | 7 | 12 | Mar 2026 | 472 |
| Ansvisor | 141 | 787 | 65 | 64 | 10 | Mar 2026 | 503 |
| Cited | 126 | 21 | 1 | 0 | 1 | Sep 2026 | 21 |
| Gego | 106 | 118 | 2 | 1 | 12 | Oct 2025 | 9 |
| Lettertrace | 75 | 323 | 5 | 4 | 1 | Jul 2026 | 323 |
We left two projects out. GetCito is a fork of Echo that has fallen behind it, and its fork count jumped in a few days in September 2026, mostly from recently created accounts with no other activity. Look through its forks and judge for yourself. GeoLook is self-host only and hasn't had a commit since August 2026.
The open-source options, tool by tool
Echo is the longest-running open-source option, released under the MIT license with broad engine coverage. The other open-source projects we track are below. Most started in 2026, and each makes real tradeoffs.
Ansvisor
FOpen-source AI search intelligence platform with a managed cloud
Limelit
FOpen-core AI visibility tracker shipped as a single self-hosted binary, with a credit-based cloud
GEO/AEO Tracker
FOpen-source, self-hosted AI visibility dashboard
Gego
FGPL-3.0 self-hosted GEO tracker with CLI and dashboard
OpenSight
FMIT-licensed AI visibility tracker with a free hosted tier
Cited
FOpen-source AI citation monitoring with a low-cost hosted cloud
Lettertrace
FMIT-licensed, bring-your-own-key AI visibility monitor
Echo
Echo covers the essential AEO workflow end to end. It tracks how ChatGPT, Google AI Mode, Google AI Overviews, Gemini, Perplexity, Copilot, Claude, and any OpenRouter model mention and cite your brand, and turns that into visibility, share of voice, sentiment, and position scores, citation and query fan-out analytics, and ranked recommendations for what to fix. It has the most polished interface of the projects here and the best agent experience: a REST API with TypeScript and Python SDKs, an MCP server, and an agent skill let Claude Code, Cursor, and Codex query your data and manage prompts directly. Agencies can white-label it, and it is MIT-licensed.
It fits teams that want to own their AEO data outright. The self-hosted core runs on Docker and PostgreSQL. You supply your own model API keys, which carry usage costs, and there is no license fee or per-seat charge. Echo does not do prompt-volume estimates, content generation, shopping-result tracking, or geographic breakdowns. It measures what AI says about you and tells you what to change, and leaves the writing to you. Managed cloud hosting is available for teams that would rather not run the infrastructure.
OneGlanse
OneGlanse is the pick if you want a completely free setup. It drives the ChatGPT, Gemini, Perplexity, Claude, and Google AI Overview web interfaces with real accounts from your own machine or a VPN, so there are no scraper or API bills, and data lands in ClickHouse on your own infrastructure.
The catch is upkeep. When those sites change their interfaces, the scrapers break until someone patches them, so it reacts slowly to changes and is harder to scale and maintain than a tool built on maintained scraping providers, like Echo. Most of its commits landed in one burst in April and May 2026, and it has never cut a tagged release.
GEO/AEO Tracker
GEO/AEO Tracker runs entirely in the browser and stores its data there, so there is nothing to deploy. It covers ChatGPT, Perplexity, Gemini, Copilot, and Google AI Mode with your own keys.
That also limits it. Your history lives in one browser on one computer, so sharing it across machines or teammates means wiring up your own Supabase project. It is a one-developer project, and the interface feels quickly assembled rather than designed.
Canonry
Canonry is built for AI agents first. Clients are configured in YAML and the work runs through a built-in agent and an MCP adapter, which suits a bot doing the work far better than a person who wants to open a dashboard and explore. It calls model APIs with your own keys rather than capturing the consumer apps. Its server-log ingestion for measuring AI-driven traffic is a genuine strength, and it is now MIT-licensed.
Ansvisor
Ansvisor lists a lot of features, from prompt tracking and citations to an Action Center, an API, and an MCP server, with a paid cloud alongside the MIT code. It is young, though: it started in March 2026, the surface changes quickly, and on its cloud the engines unlock by plan, with Claude reserved for Enterprise.
Limelit
Limelit Open covers seven engines through vendor APIs or consumer-app scrapers and has a built-in MCP server. It is open core: sentiment, prompt generation, Search Console and GA4 integration, agents, and white-label are paywalled behind the credit-based Limelit Cloud.
Self-hosting means running its Go binary or Docker, and data lives in SQLite, which is simple for one person but harder to back up, scale, and share across a team than a server database.
Cited
Cited tracks one thing: whether your verified domain is cited, mentioned, or missed for the prompts you choose. There is no sentiment, visibility scoring, share of voice, or recommendations, and it doesn't cover Copilot or Grok. Self-hosting needs DataForSEO credentials, and it is AGPL-3.0, so offering a modified version as a service means publishing your changes.
Gego
Gego is an extremely simple tracker driven from a CLI. It schedules prompts against OpenAI, Anthropic, Google, Perplexity, or local models via Ollama and records mentions and citations, with little analysis on top. A full deployment still expects PostgreSQL, MongoDB, and etcd, and it is GPL-3.0, so modified versions you distribute must stay under the GPL.
Build it yourself: scripting AI visibility checks
If no existing tool fits, the underlying job is not complicated to script. Send your prompts to the model APIs, directly or through a router like OpenRouter, then parse each response for your brand name and any links back to your site. Store the results and repeat on a schedule, because a single run is a snapshot and answers shift over time.
Everything around that loop is the work. You have to cover enough engines, handle the ones without clean APIs, keep it running, and build some way to actually read the output. That upkeep is most of what you pay for when you buy a tool, or skip by self-hosting one that already does it.
Self-hosted, open-source cloud, or closed-source cloud: the real tradeoffs
There are three ways to run AI visibility tracking, and the open-source cloud is the best of all worlds. Self-hosting gives you full control, but the upkeep is yours. A closed-source cloud takes the work off your hands, but you can't see how its numbers are made and you can't take your history with you. An open-source cloud does both: no maintenance burden, usually a lower price than closed tools, features you can contribute yourself, and the option to switch to self-hosting at any time.
| Open source, self-hosted | Open source, cloud | Closed source, cloud | |
|---|---|---|---|
| Cost | No license fee. You pay for infrastructure, AI provider APIs, and engineering time. | A subscription, usually cheaper than closed tools, with no infrastructure to run. | A subscription, often metered by prompt, seat, or engine. |
| Setup and upkeep | You deploy it and keep it running as engines and APIs change. | Sign up and start tracking. The vendor runs it and keeps engine coverage current. | Sign up and start tracking. The vendor runs it. |
| Transparency | Read the code and verify how every metric is built. | The same code as the self-hosted version, so every metric is still auditable. | The scoring is usually a black box. |
| Shaping the product | Change anything, though you maintain your changes. | Contribute features upstream and they ship for you and everyone else. | File a feature request and wait. |
| Lock-in | None. Your data is in your own database. | None. Move to self-hosting whenever you want, on the same software. | Your history lives with the vendor, and leaving means starting over. |
Echo offers both open-source options: self-host it for free, or use Echo Cloud and move to your own infrastructure whenever you want.
How to choose
Start with how you will use it rather than with a feature grid. For most teams the answer is an open-source cloud: you get a hosted product with no upkeep, auditable numbers, and a way out to self-hosting if you ever need it. Echo offers that, along with free self-hosting for teams that want full control, and it has the essential AEO features, the best UX, and the best agent experience of the projects here. If you want a fully free setup and can maintain scrapers yourself, OneGlanse is an option. If your priority is server-log attribution and your workflow is run by AI agents, Canonry goes furthest.
Weigh maintenance above features. Whether anyone will still be working on the project in a year is what the contributor column tells you, and in a field where models get renamed and APIs shift every few months, an unmaintained tracker quietly stops being correct rather than visibly breaking.
Then weigh coverage, maintenance, and your own capacity to run infrastructure against what a managed subscription would cost. If you have no time to self-host, that is useful to know early: see the best AEO tools for hosted options, or the free AI visibility tools guide for what a zero-budget setup can and cannot do. Whatever you pick, you want the same thing: a reliable, repeatable read on whether AI answers cite you, from a tool whose workings you can inspect.
Frequently Asked Questions
- What are the best open-source AEO tools?
- Echo is the best open-source answer engine optimization tool for most teams. It is MIT-licensed, has the essential AEO features, the best UX, and the best agent experience, and covers ChatGPT, Google AI Mode, Google AI Overviews, Gemini, Perplexity, Copilot, Claude, and any OpenRouter model. You can self-host it for free or use Echo Cloud. Other open-source projects exist, but each makes a significant tradeoff: browser-only storage, scrapers you maintain yourself, workflows built for agents rather than people, features paywalled behind a cloud, or citation-only tracking. GetCito is a lagging fork of Echo.
- What are the best free tools to track brand mentions in ChatGPT?
- Self-host Echo. It is MIT-licensed and free, tracks ChatGPT alongside Google AI Mode, Google AI Overviews, Gemini, Perplexity, Copilot, and Claude, and has no license fee or per-seat charge. Free trials of hosted tools work for a one-off check but stop when the trial does. You supply your own scraper or model API keys, which carry usage costs, so free means no software license rather than no spend.
- Is there a free AI visibility tool?
- Yes. Echo is free to use and free to self-host under the MIT license. The cost you cannot avoid is model and scraper API usage, because measuring visibility means actually running prompts through the engines. If you would rather not run the infrastructure, Echo Cloud runs the same software for you, and you can move to self-hosting at any time.
- Which open-source AEO tool is the most actively maintained?
- Echo. As of 17 October 2026 it has the longest continuous history of the open-source AEO tools, starting in July 2025, the most releases at 36, the most stars among actively developed independent projects, and 374 commits in the last 90 days. Many of the other projects started in 2026, and some have already gone quiet: GEO/AEO Tracker and Gego each logged 10 or fewer commits in the same window.
- Is there an open-source AI visibility tracker?
- Yes. Echo is an open-source AI visibility platform released under the MIT license. You can self-host it for free or use Echo Cloud, and because the code is open you can read exactly how every metric is built and extend it yourself.
- Is there an open-source alternative to Profound or enterprise AEO tools?
- Yes. Echo is an open-source alternative to enterprise platforms like Profound. It tracks how AI answer engines mention and cite your brand, benchmarks competitors, and exports your data, all under the MIT license with every metric computed by code you can read. Self-host it for free with your own API keys, or use Echo Cloud for a hosted version without the enterprise price.
- Should I use GetCito or Echo?
- Use Echo. GetCito is a fork of Echo: on 10 July 2026 it replaced its own codebase with Echo's, so everything it offers came from Echo. It has shipped less since, with 17 commits to Echo's 98 in the month after the fork, so it is falling further behind rather than building past it.
- Can I self-host an AEO tool?
- Yes. Echo is built to self-host: deploy it with the CLI on Docker and PostgreSQL, and your prompts and visibility history never leave your environment. You bring your own scraper and model API keys, which carry usage costs, but there is no license fee or per-seat charge. If you would rather not run it, Echo Cloud runs the same software, and you can move between the two.
- Can I build my own AI visibility tool?
- You can. The core loop is straightforward: send your prompts to the engines, parse each answer for brand mentions and citations, and log the results over time. The work is in covering enough engines, keeping scrapers working as the apps change, and running it at scale. Echo already does all of that, and because it is open source you can extend it instead of starting from scratch.
- Why use an open-source AI visibility tool?
- Because you can audit it and own it. With open source, you read exactly how each visibility metric is collected and computed instead of trusting a black-box score that might land in a board report. You self-host it, so prompts and history stay on your own infrastructure, and there is nothing to migrate off if you switch. No vendor lock-in.
- Is Echo really open source?
- Yes. Every line of Echo is open source under the MIT license and available on GitHub. You can read exactly how each metric is collected and computed, self-host it on your own infrastructure for free, and export your data at any time.
Ready to track your AI visibility?
Deploy Echo in minutes and start monitoring how ChatGPT, Claude, and Google AI Overviews talk about your brand. Open source, self-hosted, free.