AI Agents

• View as MarkdownView this page as plain text

This hub is for builders who ship Agents workflows and are looking to connect MCP clients to live chain data, submit transactions on the Rootstock network.

Core pillars

Pillar	Description	Primary tool
Connectivity	Expose chain reads and scoped actions to MCP clients.	Rootstock MCP Server
AI Agents	A lightweight dApp that connects a conversational AI agent to the Rootstock testnet, allowing users to perform DeFi actions like checking token balances and sending tRBTC simply by chatting.	AI Agent Rootstock

Key concepts

1. Model Context Protocol (MCP)

MCP connects clients to tools and data sources. The Rootstock MCP Server can read balances, contract state, and gas estimates so responses reflect current chain data.

2. Conversational Agents with Blockchain Actions

Build a lightweight dApp that connects a conversational AI agent to the Rootstock testnet, allowing users to perform DeFi actions like checking token balances and sending tRBTC simply by chatting. This is not just a chatbot—it is a minimal DeFi agent that can reason over wallet data, maintain conversational context, and issue token actions with a human-like touch.

Tools

• Rootstock MCP Server: Exposes chain reads and safe actions to MCP clients. Access the Server.
• LangChain / Eliza: Orchestration frameworks you can point at Rootstock RPC and wallets.

Implementation guides

Introduction to MCPs

This section demonstrates how to interact with the Rootstock blockchain to build and deploy innovative AI-enabled dApps.

Build Conversational AI Agents

This tutorial demonstrates how to build a lightweight dApp that connects a conversational AI agent to the Rootstock testnet, allowing users to perform DeFi actions like checking token balances and sending tRBTC simply by chatting.