GenLayer AI DAO Risk Analyzer

Minimal GenLayer demo app for AI-powered DAO proposal risk analysis with a Python intelligent contract, an Express API, and a simple frontend.

Live demo

Open the deployed app here:

https://genlayer-ai-dao-risk-analyzer.vercel.app/

Overview

This project shows how a GenLayer intelligent contract can analyze DAO governance proposals and return an AI-generated risk assessment.

What the app does

• The frontend provides a textarea for a DAO proposal and an Analyze Risk button.
• The Express backend receives the proposal text and calls a deployed GenLayer intelligent contract.
• The Python contract uses an LLM through gl.nondet.exec_prompt(...) to classify the proposal as Low, Medium, or High risk and returns a short explanation.

Project structure

.
|-- contracts/
|   `-- dao_risk_analyzer.py
|-- public/
|   |-- app.js
|   |-- index.html
|   `-- styles.css
|-- scripts/
|   `-- deploy-contract.js
|-- server/
|   |-- genlayerService.js
|   `-- index.js
|-- .env.example
|-- package.json
`-- README.md

Prerequisites

• Node.js 18+
• A running GenLayer environment such as localnet or studionet
• A deployed copy of the Python contract

Setup

1. Install dependencies:

   npm install

2. Copy the environment file and fill in the values:

   copy .env.example .env

3. Set these variables in .env:

• GENLAYER_MOCK_MODE: set to true to use the built-in local demo analyzer instead of a live contract
• GENLAYER_NETWORK: for example localnet
• GENLAYER_RPC_URL: for example http://localhost:4000/api
• GENLAYER_CONTRACT_ADDRESS: the deployed contract address
• GENLAYER_PRIVATE_KEY: only required if you want to deploy from the script or enable auto-deploy

Deploy the contract

If you already deployed the contract elsewhere, just set GENLAYER_CONTRACT_ADDRESS in .env.

To deploy this contract from the project:

npm run deploy:contract

The script prints the contract address. Copy that value into .env as GENLAYER_CONTRACT_ADDRESS.

Run locally

npm start

For local development, open http://localhost:3000 in your browser.

If you start the app without any GenLayer credentials, it automatically falls back to a local demo analyzer so the UI is still previewable.

Deploy on Vercel

This repo is Vercel-compatible:

• Files in public/ are served as the frontend
• api/health.js and api/analyze.js are deployed as Node.js Vercel Functions
• vercel.json increases the analysis function timeout and includes the Python contract source when auto-deploy is enabled

Recommended Vercel environment variables:

• GENLAYER_MOCK_MODE=true for a UI-only demo deployment
• or set GENLAYER_RPC_URL, GENLAYER_CONTRACT_ADDRESS, and any other GenLayer variables for live contract calls

How the analysis flow works

1. The browser posts proposal text to POST /api/analyze.
2. The Express server calls the deployed GenLayer contract method analyze_proposal.
3. The intelligent contract prompts an LLM and returns JSON with risk_level and explanation.
4. The backend reshapes the response into { riskLevel, explanation }.
5. The frontend renders the result.

Example response

Risk Level: Medium
Explanation:
Large treasury transfer without clear ROI.

Notes

• This demo keeps the UI intentionally simple.
• The contract is stateless, so each request returns an analysis directly instead of storing it on-chain.
• GENLAYER_AUTO_DEPLOY=true can be used with a private key if you want the server to deploy the contract automatically on first request.
• When GenLayer is not configured, the backend returns a deterministic mock analysis so the preview works without a live node.