Back to projects
2026AI & Web3liveAutomates multi-agent governance and secure treasury disbursements entirely on-chain
AgentDAO
Autonomous AI agent governance framework and live treasury control system running on Monad testnet.
Tech Stack
SolidityFoundryNext.jsReactTailwind CSSPythonLangChainGroq APIExpress.jsviem
Key Highlights
- Registers autonomous agents with specific roles and identities on-chain using an AgentRegistry contract
- Evaluates and votes on proposals programmatically using Python agent decision loops
- Integrates GovernorDAO governance logic to handle proposal timelines, timelocks, and payouts
Overview
AgentDAO is an agent-native governance platform built during the Monad Blitz hackathon. It demonstrates a fully autonomous AI-driven treasury loop: a Marketing Agent (Python, LangChain, Groq) generates and submits marketing campaign proposals, which a Finance Agent evaluates and votes on based on real-time rules (budget caps, treasury reserves, and ROI thresholds). Approved proposals flow through smart contracts (AgentRegistry, AgentTreasury, GovernorDAO) on the Monad testnet to release funds automatically, visualised on a real-time Next.js dashboard.
Project Details
- Year
- 2026
- Category
- AI & Web3
- Status
- live
- Stack size
- 10 technologies