Analysis updated 2026-05-18
Test a startup pitch or product concept against five AI personas in two minutes before presenting to real people.
Evaluate competing taglines or messaging variants to see which one generates more interest from a simulated target audience.
Interview AI personas about pain points in a product category to surface objections you had not considered.
| itsrouk/your-ai-focus-group | avacocloud/avaco-deno | doorman11991/budget-aware-mcp | |
|---|---|---|---|
| Stars | 19 | 19 | 19 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | easy | moderate | easy |
| Complexity | 2/5 | 3/5 | 3/5 |
| Audience | pm founder | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires a free Google Gemini API key from aistudio.google.com.
Your AI Focus Group is a web app that lets you test a product idea, tagline, or pitch by interviewing a panel of AI-generated personas and receiving a scored summary of what worked and what did not. The whole process takes a few minutes and runs free with a Google Gemini API key. When you open the app, you describe what you want to test. The tool rewrites your input to remove specific brand names, so the AI personas evaluate the concept and category rather than reacting to a known brand. It then generates ten synthetic personas shaped around your context. You pick three to five that match your target audience, and those become your focus group panel. You can ask up to five open-ended questions. Each persona reads and responds to each question one at a time rather than all at once, which the author designed to feel more like a real conversation. After the interview, each response receives a score on a 1-to-5 scale using a method called follow-up Likert rating. You get a scorecard that shows which responses landed well, which drew pushback or low interest, how much the panel agreed across responses, and the recurring themes from their answers. You can also copy the full transcript as plain text. The app is built with Next.js 14 and TypeScript, uses Tailwind CSS for styling, and calls Google Gemini 2.5 Flash to generate the personas and their responses. State is managed with React's built-in useReducer without any external state library. To run it locally, you clone the repository, install dependencies with pnpm, add a free Gemini API key to a .env.local file, and start the dev server. The project is licensed under MIT.
A web app that generates AI personas for your product idea, interviews them with your questions, and gives back a Likert-scored summary of what resonated.
Mainly TypeScript. The stack also includes TypeScript, Next.js, Tailwind CSS.
MIT license: use, modify, and distribute freely for any purpose including commercial use.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly pm founder.
This repo across BitVibe Labs
Verify against the repo before relying on details.