Convert smart contract ABIs into AI-ready MCP servers that can read contract state and execute transactions. Built for developers integrating blockchain interactions into AI agents and LLM workflows.
io.github.nirholas/abi-to-mcp
Local install
STDIO
No auth required
How models use it and what it is built for.
Convert smart contract ABIs into AI-ready MCP servers that can read contract state and execute transactions. Built for developers integrating blockchain interactions into AI agents and LLM workflows.
Local install — runs as a subprocess.
Configuration this server reads at startup.
Ethereum JSON-RPC endpoint URL
Etherscan API key for fetching verified contract ABIs
Private key for signing transactions (write operations)
Where to find authoritative docs and source for UCAI.
Paste any of these into Agent Studio after connecting UCAI.
Common questions about connecting and running UCAI.
What does abi-to-mcp actually do?
It transforms a smart contract ABI (Application Binary Interface) into an MCP server, making contract functions callable through an AI chat interface. You provide an ABI and configuration, and it generates a server that can read state and execute transactions.
How do I fetch a contract ABI automatically?
Set the ETHERSCAN_API_KEY environment variable with your Etherscan API key. The server can then fetch verified contract ABIs directly from Etherscan instead of requiring you to paste them manually.
Can I use this to send transactions?
Yes, if you provide a PRIVATE_KEY environment variable, the generated MCP server can sign and execute write operations on the contract. Always use a key with limited funds for testing.
What RPC endpoint should I use?
Set WEB3_PROVIDER_URL to any Ethereum-compatible JSON-RPC endpoint (Infura, Alchemy, Arbitrum, Polygon, etc.). This determines which blockchain the server interacts with.
Is this different from calling a contract directly in code?
Yes—this wraps contract ABIs as MCP tools, letting LLMs and AI agents discover and call contract functions naturally through conversation, rather than requiring custom code for each integration.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent StudioJudges Panel
45 judges that evaluate AI-generated code for security, cost, and quality with built-in AST.
Addozhang Nexus
Query Sonatype Nexus Repository (OSS/Pro) for Maven, Python, and Docker artifacts
Cyanheads MCP Ts Template
TypeScript template for building MCP servers with declarative tooling, observability, and auth.
Chernistry Bernstein
Declarative agent orchestration for engineering teams