mcpField
B
Blocknative MCP Server

Blocknative MCP Server

Local
github.com/kukapayFinance4Updated 78w ago

About Blocknative MCP Server

Blocknative MCP Server exposes Blocknative gas-price predictions and transaction-cost estimates for developers and MCP clients working across supported blockchains. It also lists supported chains and provides a gas-query prompt.

Key features

  1. 01Gas-price predictions by chain
  2. 02Transaction-cost estimates in Gwei and ETH
  3. 03Supported-chain listing
  4. 04Gas-query prompt template

Use cases

  1. 01Check fee estimates before submitting a transaction
  2. 02Compare gas-price confidence levels across chains
  3. 03Find supported chain IDs and networks

MCP support

Exposes gas prediction, cost estimation, and supported-chain tools, plus a gas-price prompt, to MCP clients.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Blocknative MCP Server

An MCP server that provides real-time gas price predictions across multiple blockchains, powered by Blocknative.

GitHub License Python Version Status

Features

  • Tools:
    • predict_gas_price: Fetches gas price predictions for a specified chain (default: Ethereum Mainnet, chain_id=1), including base fee and a table with confidence levels, price, max priority fee, and max fee.
    • estimate_gas_cost: Estimates transaction costs based on gas limit, confidence level, and chain ID, returning costs in Gwei and ETH.
    • get_supported_chains: Lists supported blockchains in a table with chain ID, system, and network.
  • Prompt:
    • gas_price_query: A prompt template for querying gas prices at a specific confidence level and chain ID.
  • Asynchronous: Uses httpx for non-blocking HTTP requests to Blocknative's Gas Price and Chains APIs.
  • Optional API Key: Supports low-frequency access without a Blocknative API key; high-frequency use requires setting BLOCKNATIVE_API_KEY.

Prerequisites

  • Python: Version 3.10
  • uv: For dependency management and running the project (installation guide)
  • Blocknative API Key (optional): Required for high-frequency API access. Sign up at Blocknative to obtain a free API key.

Installation

  1. Clone the Repository:

    git clone https://github.com/kukapay/blocknative-mcp.git
    cd blocknative-mcp
    
  2. Set Up the Project with uv:

    uv sync
    
  3. Set the Blocknative API Key (Optional):

    • For high-frequency access, set the environment variable:
      export BLOCKNATIVE_API_KEY="your-api-key-here"
      
    • For low-frequency access, skip this step.

Usage

The server provides three tools and one prompt, accessible via the FastMCP framework. You can run the server in development mode, execute it directly, or integrate it with Claude Desktop.

Running the Server
  1. Development Mode with MCP Inspector:

    uv run mcp dev blocknative-mcp
    

    This opens the MCP Inspector, a web interface for testing tools and prompts interactively.

  2. Direct Execution:

    uv run blocknative-mcp
    
  3. Claude Desktop Integration:

    uv run mcp install blocknative-mcp --name "Blocknative MCP Server"
    

    This installs the server for use with Claude Desktop.

Tools
  • predict_gas_price(chain_id: int = 1) -> str: Fetches gas price predictions for a chain (default: Ethereum Mainnet). Returns base fee and a Markdown table with confidence levels, price, max priority fee, and max fee.

    • Example:
      Gas Price Predictions for Chain ID 1 (ethereum/main):
      - Base Fee Per Gas: 0.382910791 Gwei
      
      | Confidence | Price (Gwei) | Max Priority Fee (Gwei) | Max Fee (Gwei) |
      |------------|--------------|-------------------------|----------------|
      | 99%        | 0.52         | 0.14                    | 0.9            |
      | 95%        | 0.48         | 0.094                   | 0.86           |
      | 90%        | 0.47         | 0.089                   | 0.85           |
      | 80%        | 0.46         | 0.079                   | 0.84           |
      | 70%        | 0.45         | 0.069                   | 0.83           |
      
  • estimate_gas_cost(gas_limit: int, confidence: int = 99, chain_id: int = 1) -> str: Estimates transaction costs based on gas limit, confidence level, and chain ID. Returns costs in Gwei and ETH.

    • Example:
      Estimated Gas Cost (Confidence 90%, Chain ID 1):
      - Gas Limit: 21000
      - Max Fee Per Gas: 0.85 Gwei
      - Total Cost: 17850 Gwei (0.00001785 ETH)
      
  • get_supported_chains(ctx: Optional[Context] = None) -> str: Lists supported chains in a Markdown table with chain ID, system, and network.

    • Example:
      Supported Chains:
      
      | Chain ID   | System   | Network   |
      |------------|----------|-----------|
      | 1          | ethereum | main      |
      | 137        | polygon  | mainnet   |
      | 8453       | base     | mainnet   |
      | 11155111   | ethereum | sepolia   |
      
Prompts
  • gas_price_query(confidence: int, chain_id: int = 1) -> List[base.Message]: A prompt template for querying gas prices at a specific confidence level and chain ID.
    • Example:
      - User: What is the current gas price for chain ID 1 transactions with 90% confidence?
      - Assistant: Let me fetch the gas price predictions for chain ID 1.
      

License

This project is licensed under the MIT License. See the LICENSE file for details.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.