Jolly LLM is a retro-styled AI courtroom simulator that generates both Prosecution and Defense arguments for any real-world legal or ethical prompt.
Built in 24 hours at the Pixel Palettes Hackathon by team Prison Breakers, the app blends pixel aesthetics, fast LLMs, and theatrical court logic into a one-of-a-kind experience.
🟢 Frontend (Netlify):
👉 https://jolly-llm-pixel.netlify.app
🟢 Backend API (Render):
👉 https://jolly-llm.onrender.com/api/generate
🔧 GitHub Repo:
👉 https://github.com/EmperorsReign05/jolly-llm
- User enters a case-like prompt (e.g., "A student was caught cheating in an entrance exam").
- The system sends it to an LLM (via Groq API using LLaMA 3).
- The backend returns detailed arguments for both:
- 🟥 Prosecution
- 🟩 Defense
- Text is animated in retro typewriter style, with glowing pixel borders and terminal UI.
- 🧑🎓 Law & ethics education
- 🧠 Debate prep
- ⚖️ Thought experiments
- 🎮 Fun & theatrical roleplay with AI
- Frontend: Vite, HTML, CSS, JavaScript
- Backend: Node.js, Express.js
- LLM: Groq API (LLaMA 3)
- Deployment: Netlify (frontend), Render (backend)
- Styling: Press Start 2P font, pixel court visuals, typewriter effects
# Clone the repo
git clone https://github.com/EmperorsReign05/jolly-llm
cd jolly-llm
# Install backend deps
npm install
# Run backend
node server.js
# In another terminal, run frontend
npm run dev