Overview
The Bankless Onchain MCP Server is an MCP-compatible server for blockchain data interaction through the Bankless API. It provides a framework for interacting with on-chain data via the Bankless API and implements the Model Context Protocol (MCP) to allow AI models to access blockchain state and event data in a structured way. The server exposes a variety of on-chain data operations, including contract state reading, proxy resolution, ABI retrieval, source retrieval, event querying, event topic construction, transaction history, and transaction details. Supported tools include read_contract, get_proxy, get_events, and build_event_topic, each with clearly defined inputs (network, addresses, methods, topics, inputs/outputs) and outputs (typed values, ABIs, event logs, topic hashes, etc.). Installation is via npm install @bankless/onchain-mcp and running with npx @bankless/onchain-mcp, after setting BANKLESS_API_TOKEN. It is designed to be used as an MCP tool provider for AI models, with usage examples for each tool and guidance on integration with LLMs. Error handling covers BanklessValidationError, BanklessAuthenticationError, BanklessResourceNotFoundError, and BanklessRateLimitError. Version: 0.6.2.
Features
Read Contract State
Read state from smart contracts on various blockchain networks.
Get Proxy
Retrieve proxy implementation contract addresses.
Get ABI
Fetch the ABI for a contract in JSON format.
Get Source
Retrieve verified contract source code, ABI, compiler version, and metadata.
Get Events
Fetch event logs for a contract based on topics.
Build Event Topic
Generate an event topic signature from event name and argument types.
Get Transaction History
Retrieve transaction history for a user address with optional filters.
Get Transaction Info
Get detailed information about a specific transaction.
Who Is This For?
- AI models:Access blockchain state and event data via MCP for AI reasoning and data-driven analysis.
- Blockchain developers:Integrate MCP tools to read contract state, events, and transactions for applications.
