Built with the PR community, for the PR community
AI Visibility, open-sourced
Track how your brand appears across ChatGPT, Gemini, Claude, Perplexity and Grok — and validate whether what they say is actually correct.
We've taken the AI Visibility module of the Agentcy platform and made it completely free and open for anyone to download, run, inspect and build upon. No proprietary score. No black box. A transparent, reproducible method you can stand behind with clients.
Announced at the AI for PR Conference, London — June 2026.
$ git clone github.com/AgentcyPR/ai-visibility.git
$ cd ai-visibility
$ bash setup.sh
✓ installed dependencies
✓ added your AI provider keys
✓ created local database + demo data
$ npm run dev
▸ running at http://localhost:3000
For PR teams, presence is no longer enough. What the AI says about you matters as much as whether it mentions you at all. A brand can appear constantly in AI answers and still be systematically misrepresented — or cited as a source without ever being named in the response.
What the platform measures
Run prompts against multiple models and get a full read on how your brand — and your competitors — show up in AI answers.
Visibility
How often a brand is mentioned or recommended across the leading AI models.
Accuracy
Whether those mentions are factually correct and well positioned, validated claim-by-claim against a brand source of truth.
Citations
Which sources AI models cite in their answers — and whether those citations actually translate into brand mentions.
Sentiment
Whether mentions are positive, neutral or negative in tone.
Competitor benchmarking
How rivals are positioned in the very same responses about your category.
Trends over time
How visibility, accuracy and positioning shift across scheduled, repeatable runs.
Visibility tells you if you're in the room. Accuracy tells you whether what's being said is true.
Visibility tracking, citation analysis, sentiment, competitor benchmarking and trend monitoring are now table stakes. What's been missing is an accuracy layer: a way to validate whether what AI says about your brand is factually correct.
The platform breaks each brand description into discrete, testable claims, checks them against approved brand material, and flags every divergence — wrong facts, outdated positioning, missing context, competitive misattribution. Sentiment is not accuracy. A cited domain is not proof the answer is right.
Every score is fully traceable
- The raw prompt that was asked
- The raw answer the model returned
- The exact URLs the model cited
- A timestamp for every run
- Each claim checked against your source of truth
Nothing is a black box — because you're running it yourself.
Why we open-sourced it
The PR community needs to own AI visibility as an important communications channel for reaching its target audience. AI answers are fast becoming where reputations are made and lost — and that infrastructure shouldn't sit behind a paywall or a score nobody can inspect.
The market is already pushing back. A recent Digiday investigation (1 May 2026) found marketers questioning expensive AI visibility tools after inconsistent results and a limited ability to prevent the hallucinations and misattributions they were bought to address.
By making the measurement layer open source, we're giving agencies a foundation they can inspect, verify and run themselves — a transparent, reproducible method rather than a proprietary number. It was built with the industry and for the industry, with input from senior practitioners across PR, corporate communications and public affairs. If PR is going to be taken seriously as a performance discipline, accuracy has to be part of what we measure.
Much of the hype about GEO centres on visibility, but for the work we do with clients in the corporate, government and not-for-profit space, being portrayed accurately in AI answers is even more important. Tools that just focus on visibility can't give corporate communications professionals the data they need to protect and improve reputations.
Agencies are paying for tools that measure whether a brand appears in a result — not whether what it says is correct. That's the gap we built this to address. Open-sourcing the measurement layer means any agency can verify the results themselves, without trusting a black box they didn't build and can't inspect.
Try it on your own machine in minutes
Run the entire app locally with SQLite — no Docker, no cloud database. The only cost is the AI tokens you use. When you're ready to offer access to clients, switch to PostgreSQL and deploy to your own infrastructure.
- Bring your own keys. Connect the providers you choose — OpenAI, Google, Anthropic, Perplexity or xAI.
- Self-hostable. SQLite for free local testing, PostgreSQL for production. Deploy with Docker on your own infrastructure.
- Yours to build on. MIT-licensed, with full documentation and setup instructions included.
# 1. Get the code
$ git clone github.com/AgentcyPR/ai-visibility.git
$ cd ai-visibility
# 2. One-command setup (adds your keys + demo data)
$ bash setup.sh
# 3. Start the app + worker together
$ npm run dev
▸ open http://localhost:3000
sign in: [email protected] / password
Requires Node.js 18+ and Python 3.10+ (for the background worker) and at least one AI provider API key.
Own your AI visibility
Released under the MIT licence and available now. Clone it, run it, inspect it, and make it part of how your agency measures the AI channel.