Cheap Seats: an open-source, self-hosted AI visibility tracker
Find out whether AI assistants recommend you, at no cost. Cheap Seats is an open-source project I wrote to make my own product better, drawing on lots of inspiration, Gauge first among them. It's free, and always will be: no hosted tier, no upsell. It runs on the Claude subscription you already have (ChatGPT optional), asks your buyers' questions, grades whether each answer recommends your brand, merely mentions it, or leaves it out, and builds a ranked list of fixes.
MIT licensed, code at github.com/cheapseatsdev/cheapseats. Your prompts live in your own git repo, so if you outgrow Cheap Seats, they come with you.
How it works
Your buyers' questions
Prompt sets are TOML files in a config repo you own, one per category of buyer question, written the way someone who has never heard of you would type them. Changes go through pull requests.
Two tracks, every assistant
Each configured assistant answers every prompt on search, where it must search the web first, and recall, with no tools. You see both what the web says today and what the model already believes.
Graded answers
A second Claude call grades each answer
recommended,mentionedorabsent, with your position, the top pick, every product named, the objections raised, and why you weren't #1.Share of voice and rankings
Each run shows rates per assistant and track with 95% intervals, and share of voice. Rankings follow your place day by day, and carry on through reworded, added and removed prompts.
A ranked fix list
Each run ends with Fix these first: the exact page to create or change, what it must say, the prompts it would flip, an effort estimate, and which of your sites it belongs on, with its owner.
Start here, move on when you're ready
I built Cheap Seats to make my own product better. It draws on a lot of tools, Gauge first among them, along with Profound, Peec AI and others. They do a lot, and most of it I didn't need; what I did have was Claude and ChatGPT subscriptions with tokens to spare. So Cheap Seats does the part you'll use most: ask your buyers' questions, grade the answers, show which pages shape them, and say what to fix first. It runs on those subscriptions, on your own machine, with nothing to buy. It's free, and it always will be.
It's for the people who own how a product is described: product marketing, docs, the blog and the website, and anyone starting to measure AI visibility. I hope you find it useful. If you grow out of it, because you need engines it doesn't ask, a hosted service, or reporting for many clients, do whatever suits you. I'd suggest Gauge; that team is great. The buyer questions you've refined are plain text in your own repo, ready to take with you. How the tools compare.
Open source, on your own infrastructure
- MIT licensed. Read it, run it, change it.
- Self-hosted. One container with a SQLite database, on a Docker host, Kubernetes (Helm chart) or an exe.dev VM.
- Your config in your git repo. The brand, assistants, budget, schedule and prompt sets live in a config repo you can keep private.
cheapseats planshows what a change does to your trends before you merge it. - Your subscriptions, or the APIs. Answers, grades and the fix list can all come from the
claudeandcodexCLIs on your Claude and ChatGPT subscriptions. Through the APIs, a daily budget caps spend. Prompts and answers go only to the model providers you configure. - Web UI, CLI, JSON API and MCP server, so Claude Code or any MCP client can start runs and read results.
Get started
A fresh checkout runs against an example config that measures Cheap Seats itself. You need Go and Claude: the claude CLI logged in to a Claude subscription (no API key, no spend), or an Anthropic API key. ChatGPT is optional: the codex CLI or an OpenAI key adds its answers.
CHEAPSEATS_CONFIG=examples/cheapseats-config/cheapseats.subscriptions.toml go run ./cmd/cheapseats serve
# then open http://127.0.0.1:3000Read Getting started, then start your own config repo from the template.