Thirdweb AI

Provides multi-component MCP server for on-chain data, wallet and contract ops, cloud engine, storage, and Nebula-based AI interactions.

python

18

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "thirdweb-dev-ai": {
      "command": "uvx",
      "args": [
        "thirdweb-mcp"
      ],
      "env": {
        "THIRDWEB_SECRET_KEY": "YOUR_SECRET_KEY"
      }
    }
  }
}

You can run and manage the Thirdweb AI MCP server to access on-chain data analysis, wallet and contract operations, decentralized storage, and natural language powered blockchain interactions. This MCP server lets you start local tooling or connect to remote endpoints, enabling practical, multi-component blockchain automation and AI-assisted workflows.

How to use

Install and run the MCP server using your preferred method. You can start a local runtime that executes the MCP host and then connect your client applications to it, or run a cloud-based process that communicates with the same endpoints. After starting, you will have access to data insight, wallet and contract operations, cloud engine services, decentralized storage, and Nebula-powered natural language interactions. Use the provided tools to build AI agents that analyze chains, read and write contracts, manage server wallets, upload to IPFS, and orchestrate tasks through natural language.

How to install

Prerequisites: ensure you have a runtime for your chosen execution method. You can use the uvx runtime or install via pipx for Python-based execution. Prepare your secret key for authentication when necessary.

Option 1: Run with uvx (recommended for quick start) set your secret key in the environment and start the MCP server with the uvx runner.

Option 2: Install with pipx and run the MCP server as a Python entry point.

mcp server configuration and start commands

{
  "mcpServers": [
    {
      "name": "thirdweb_ai",
      "type": "stdio",
      "command": "uvx",
      "args": ["thirdweb-mcp"],
      "env": [
        {"name": "THIRDWEB_SECRET_KEY", "value": "YOUR_SECRET_KEY"}
      ]
    },
    {
      "name": "thirdweb_mcp",
      "type": "stdio",
      "command": "thirdweb-mcp",
      "args": [],
      "env": [
        {"name": "THIRDWEB_SECRET_KEY", "value": "YOUR_SECRET_KEY"}
      ]
    }
  ]
}

Additional setup notes

Environment variables shown here are required when you want to run the MCP server locally. Replace YOUR_SECRET_KEY with your actual secret key before starting the server. The two startup options give you a choice between the uvx-based launcher and a direct Python-based MCP runner.

Security and usage notes

Keep your secret key secure and do not expose it in client-side code or logs. If you discover a vulnerability, report it via the designated security channel and avoid using public issue trackers for security-related details.

Publishing and updates

When publishing new MCP server versions, create a release tag and push it to your hosting environment or CI/CD pipeline so that clients can fetch the latest server builds. Follow your standard release process to ensure version parity across client integrations.

Available tools

insight

Comprehensive blockchain data intelligence including chains, transactions, blocks, events, prices, and tokens.

engine

Core blockchain interaction capabilities including Wallet management, Read operations for contracts, and Write transactions.

enginecloud

Cloud-based engine operations for server wallets, contract interaction, transaction management, and balance queries.

storage

Decentralized storage to upload files, directories, and JSON to IPFS and fetch content via IPFS gateway.

nebula

AI agent capability to perform blockchain tasks through natural language instructions.