EIPs

An MCP server for providing semantically related Ethereum Improvement Proposals (EIPs) to AI agents.

python

2

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": {
    "kukapay-eips-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/eips-mcp",
        "run",
        "main.py"
      ]
    }
  }
}

You run an MCP server that serves semantically related Ethereum Improvement Proposals (EIPs) to AI agents. It provides a fast semantic search over EIP documents and processes Markdown content for efficient storage and retrieval, enabling you to quickly locate relevant EIPs for your use cases.

How to use

You interact with the server through an MCP client and use the single available tool named search. Provide natural language prompts to find relevant EIP content. The tool returns up to five relevant EIP document chunks, separated for clarity, so you can review specific sections like abstracts or motivations.

How to install

Prerequisites: Python 3.10+ and uv (recommended) or pip for dependency management.

Setup steps to get the server up and running include these commands:

# 1. Clone the repository with submodules
git clone --recursive-submodules https://github.com/kukapay/eips-mcp.git
cd eips-mcp

# 2. Install the dependencies (recommended)
uv sync

# 3. Load EIP documents
uv run load_eips.py

# 4. Install the MCP server to Claude Desktop
uv run mcp install main.py --name "EIPs"

{
   "mcpServers": {
       "EIPs": {
           "command": "uv",
           "args": [ "--directory", "/path/to/eips-mcp", "run", "main.py" ]
       }
   }
}

Replace /path/to/eips-mcp with your actual installation path.

## Additional sections

Notes on usage and configuration:
- The server exposes a single MCP tool named `search` for semantic queries over EIPs.
- When you run the `search` tool, you can ask questions like “Search for EIPs related to CREATE2” or “Find EIPs about contract deployment opcodes.”
- Results are presented as chunks from EIP documents, up to five items, with clear separators so you can review each relevant piece.

## Available tools

### search

A single MCP tool that accepts a natural language query and returns up to five relevant EIP document chunks, enabling semantic search over EIP content.