Succinct (Flappy) Bird Zero-Knowledge Proof API

This project implements a REST API using Axum and SP1 SDK to generate zero-knowledge proofs for a Flappy Bird game simulation. It processes game inputs (flaps and username) and returns a Groth16 proof with score and username verification.

---

Game Play and Prove Operation

---

🚀 Features

• Game Simulation: Simulates Flappy Bird gameplay in program.rs using flaps input.
• Proof Generation: Generates a Groth16 proof with SP1 SDK, committing score and username.
• API: Axum server at 0.0.0.0:3000/api/prove-score returns proof, public inputs, verifying key hash, score, and username in hex format.
• Docker Integration: Uses Docker for Groth16 proof generation.

---

🛠 Setup

1️⃣ Install Dependencies

cargo build --release
sudo apt install -y docker.io
sudo systemctl start docker

2️⃣ Run the API Server

RUST_LOG=info cargo run --release --bin main

---

📡 Usage

Send a POST request:

fetch('http://localhost:3000/api/prove-score', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ flaps: [39, 90, 110], username: "player1" })
}).then(res => res.json()).then(data => console.log(data));

---

📜 Response Example

{
  "proof": "11b6a09d...",
  "public_inputs": "01000000...",
  "vkey_hash": "0x005c3cf5...",
  "mode": "groth16",
  "score": 1,
  "username": "player1"
}

---

⚠ Notes

• Groth16 proof generation takes ~42s due to SNARK wrapping.
• Requires Docker for Gnark backend.

Its based on Succinct Floppy GPU

Succinct SP1 Game