Skip to content

Tre-AG/leap

Repository files navigation

Leap

AI recommendation tool — taketheleap.ai

Leap is a fun, interactive experience that helps everyday people discover how AI can make their lives easier. Instead of assuming people already care about AI, Leap surfaces their real-life friction points and maps AI solutions to them. The magic moment is: "Wait, AI can help with that?"

What It Is

A 5-minute interactive quiz disguised as a mini-game. Users pick a character class, play through scenario-based choices about their daily life, and receive personalized AI tool recommendations based on their answers.

Think 16Personalities meets Telltale Games, but for AI discovery.

Character Classes

  • The Student — Learning, studying, building skills
  • The Professional — Working a job, building a career
  • The Entrepreneur — Side hustles, projects, your own path
  • The Parent — Juggling kids, work, and everything else
  • The Creative — Making things, expressing ideas

Current Status

Planning phase complete. Build phase starting.

What's done:

  • 40 scenarios across 5 character classes
  • 15 flavor moments (story beats between scenarios)
  • 5 interactive data collection screens
  • AI prompt template tested and working
  • Tool knowledge layer (16 vetted tools with honest context)
  • Character illustrations finalized
  • Data architecture (modular JSON files, not hardcoded)
  • Visual design direction locked in

What's next:

  • Initialize Next.js project
  • Build the frontend experience
  • Integrate Claude API for personalized recommendations
  • Deploy to taketheleap.ai via Vercel

Tech Stack

  • Frontend: Next.js + React
  • Styling: Tailwind CSS
  • AI Engine: Claude API (Anthropic)
  • Hosting: Vercel
  • Domain: taketheleap.ai

Project Structure

leap/
  assets/          # Character illustrations (5 frog PNGs)
  data/
    scenarios/     # 5 JSON files (one per character class)
    flavors.json   # 15 flavor moments
    interactive.json # Tool grid, responsibilities, time drains, wishlist
    tools.json     # 16 vetted AI tools with recommendations context
  docs/
    leap-project-plan.md
    leap-tool-knowledge.md

Philosophy

  • Meet people where they are. No jargon.
  • Fun over functional. It should feel like a game, not a survey.
  • Show, don't sell. Recommendations are genuine.
  • Lowest hanging fruit first. Start with the simplest wins.
  • Depth over breadth. One tool used well beats five tools installed.

About

AI recommendation tool - taketheleap.ai

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors