AgentFi

🗄️ Archived — 2026-05-17

This project is archived. The code is functional and the npm package remains published, but the project is no longer maintained.

Honest postmortem. AgentFi reached technical MVP early (agent registration → on-chain transaction with operator fee collection worked) and then accumulated breadth (4 chains, 31 MCP tools, A2A jobs, escrow v3, revenue sharing) without ever finding a user. Over ~50 days the surface area grew while real adoption stayed at zero.

Why archived. The thesis (AI agents executing autonomous DeFi operations) remains directionally correct, but the market is ~2-3 years early — confidence in LLM-driven financial autonomy is the bottleneck, not tooling. When the market opens, well-capitalized incumbents (Coinbase Agent Kit, Safe modules, Anthropic-native primitives) will move in. No defensible moat for a solo OSS protocol in that window.

What's still useful here.

• Reference implementation of an MCP server with 31 DeFi tools.

• Working examples of Safe v7 protocol-kit integration, Uniswap Routing API v2 wrapper, Aave/Compound/GMX adapters.

• Apache-2.0 — fork freely.

Live infrastructure (Fly.io backend, Upstash Redis) was decommissioned on archive. The maintainer's smart contracts on Base Mainnet remain deployed but unused.

The economic layer for non-human intelligence.

AgentFi provides crypto transaction infrastructure for AI agents on Ethereum and EVM-compatible networks. It allows agents to execute DeFi transactions (swaps, yield farming, transfers) without handling private keys or managing gas, all within a secure on-chain policy framework.

Start here:

• VISION.md — why this project exists, where it's going, and the principles behind every technical decision.

• STATE.md — what the project is today: purpose, full stack, capabilities, phase progress.

• HANDOFF.md — live pending tasks, credentials inventory, new-machine setup.

Related MCP server: Universal Crypto MCP

📚 Documentation

All project documentation is organized in our Documentation Hub.

Quick Links

• Vision (required reading): Why this project exists and where it's going.

• Dev Quickstart: Zero-credential local stack — docker compose up and you're running in 3 minutes.

• Operator Setup: Get a real instance of AgentFi running.

• Graphify Code Graph: Queryable codebase map for agent-assisted development.

• Agent Quickstart: Connect your agent in < 5 minutes.

• System Architecture: Understand the 4-layer stack.

🚀 Key Features

• Turnkey MPC Wallets — keys split across shards and never exposed.

• Safe Smart Wallets — per-agent on-chain policy enforcement (limits, whitelists, kill switch).

• Model Context Protocol — 31 tools in @agent_fi/mcp-server: DeFi execution, GMX perpetuals, A2A collaboration, trust, and P&L.

• DeFi coverage — Uniswap V3 + Curve StableSwap (swaps); Aave V3, Compound V3, and any ERC-4626 vault (yield).

• Agent-to-Agent economy — job queue, atomic payments, DB-level escrow (v2), reputation scoring from real metrics with time-decay.

• Agent P&L dashboard — per-agent breakeven detection, including real gas costs (v2).

• Persistent identity — optional ENS subdomains (alice-abc123.agentfi.eth) wired into agent registration.

• OpenAPI 3.0.3 spec at docs/api/openapi.yaml — machine-readable contract for SDK generation.

• Cross-chain — Ethereum, Base, Arbitrum, Polygon (Base Mainnet has maintainer-deployed contracts).

• Protocol fee engine — basis-point fee collected on-chain atomically via AgentExecutor.

🛠️ Getting Started

For Operators

1. Follow the Operator Setup Checklist to fill third-party accounts and local .env.

2. Deploy the Smart Contracts to your target chain (or reuse the maintainer-deployed Base addresses in STATE.md).

3. Deploy the backend via Self-Hosted Production Guide — provider-agnostic, with Railway as the reference and Fly.io / Render / Docker documented as alternatives.

For Developers

1. Start with the Dev Quickstart — docker compose up → stack running in 3 minutes, zero external accounts.

2. Run the A2A Collaboration Example — two-agent end-to-end flow in one file.

3. Run the Claude Desktop MCP Demo — two MCP identities, discovery, A2A job, trust, and P&L.

4. Review the Architecture Overview.

5. Use the MCP Server to integrate your agents.

🛡️ Security

AgentFi is built for security-first autonomy.

• Simulations: Every transaction is simulated via Tenderly before submission.

• Guardrails: On-chain policies prevent agents from exceeding predefined limits.

• Kill Switch: Operators can pause any agent's transaction ability instantly.

Found a vulnerability? See SECURITY.md for our disclosure policy.

🤝 Community

• Code of Conduct: Our community standards.

• Contributing: How to contribute to AgentFi.

• Changelog: Release history and notable changes.

📄 License

AgentFi is open-source and licensed under the Apache 2.0 License.