PumpSwap MCP Server
@kukapay
An MCP server that enables AI agents to interact with PumpSwap for real-time token swaps and automated on-chain trading.
Overview
What is PumpSwap MCP Server?
A Model Context Protocol (MCP) server that enables AI agents to interact with PumpSwap for real-time token swaps and automated on‑chain trading on Solana.
How to use PumpSwap MCP Server?
Clone the repository, install uv, run uv sync to install dependencies, create a .env.private file with a Solana RPC endpoint, slippage settings, priority fee, and your Solana private key, then start the server with uv run main.py.
Key features of PumpSwap MCP Server
- Buy tokens with SOL using configurable slippage and priority fees.
- Sell tokens for SOL with configurable parameters.
- Retrieve current token prices in SOL.
- Fetch and display detailed pool information for a given token mint.
Use cases of PumpSwap MCP Server
- An AI agent buys tokens automatically based on market signals.
- An agent sells tokens to realize profits or cut losses.
- A bot queries token prices and pool data for analysis or decision-making.
FAQ from PumpSwap MCP Server
What does PumpSwap MCP Server do?
It exposes four MCP tools that let AI agents buy tokens, sell tokens, get token prices, and retrieve pool data on PumpSwap, all via Solana mainnet.
What are the runtime requirements?
Python 3.13 or higher, the uv package manager, a Solana RPC endpoint (default https://api.mainnet-beta.solana.com), and a valid Solana private key for transaction signing.
How do I configure the server?
Set environment variables in .env.private: HTTPS_RPC_ENDPOINT, BUY_SLIPPAGE, SELL_SLIPPAGE, SWAP_PRIORITY_FEE, and the required PRIVATE_KEY. The server will error if PRIVATE_KEY is missing.
Where does the data come from?
All token prices, pool data, and transaction results come directly from the Solana mainnet via the configured RPC endpoint. No data is stored locally.
How are transactions signed and secured?
The server uses the private key provided in PRIVATE_KEY to sign each transaction. The .env.private file must be kept secure and never shared. The project carries a disclaimer that the authors are not responsible for financial losses or security issues.
