Sakshya ZKP Circuits

This directory contains the Zero-Knowledge Proof (ZKP) circuits for the Sakshya Protocol. These circuits enable privacy-preserving verification of user attributes like age and biometric similarity without revealing the underlying sensitive data.

Overview

Sakshya leverages zk-SNARKs (via circom and snarkjs) to prove that:

1. A user is over the legal age (18+).
2. A live biometric capture matches the registered identity (score >= 95%).

The proofs are generated locally by the user and verified on-chain or by the TEE Gateway, ensuring that raw biometric data never leaves the user's secure environment.

Project Structure

• circuits/: Core logic defined in Circom.
  • age_check.circom: Logic for age verification.
  • score_check.circom: Logic for biometric score thresholding.
  • kyc_verifier.circom: Main entry point aggregating all checks.
• scripts/: Helper scripts for the ZKP lifecycle.
  • setup.sh: Compiles circuits and performs the Trusted Setup (Groth16).
  • prove.js: Node.js script to generate a proof from inputs.
  • verify.js: Node.js script to verify a generated proof.
• test/: Test suites for validating circuit logic.

Prerequisites

• Node.js (v16+)
• Circom (v2.0.0+)
• snarkjs (npm install -g snarkjs)

Getting Started

1. Installation

Install dependencies:

npm install

2. Trusted Setup

Run the setup script to compile the circuits and generate the necessary keys (ptau, zkey, vkey):

./scripts/setup.sh

3. Generate a Proof

You can generate a proof using scripts/prove.js. You can modify the sampleInput in the script or import the prove function into your application.

node scripts/prove.js

This will output proof.json and public.json.

4. Verify a Proof

Verify the generated proof against the verification key:

node scripts/verify.js

Circuit Details

KYC Verifier (kyc_verifier.circom)

The main circuit takes the following inputs:

• Public Inputs:
  • userPubKeyHash: A hash of the user's public key to bind the proof to a specific identity.
• Private Inputs:
  • age: The user's actual age.
  • biometricScore: The similarity score (0-100) from the FHE/TEE matching process.

Constraints:

• age >= 18
• biometricScore >= 95
• verified must be 1

Security Considerations

• Trusted Setup: The current setup.sh uses a simple Phase 1 contribution. For production, a multi-party computation (MPC) ceremony is required.
• Input Validation: Ensure that the inputs provided to the circuits are correctly derived from trusted sources (e.g., identity documents, TEE outputs).