AgentiPy is a Python toolkit designed to enable AI agents to interact with blockchain applications, primarily on Solana and Base. It simplifies dApp development by providing tools for token management, NFT handling, and DeFi interactions, allowing for AI-driven on-chain workflows.

How It Works

AgentiPy leverages asynchronous programming for efficient blockchain interactions and offers a comprehensive toolset for various on-chain actions. It supports a wide range of protocols, including Jupiter, PumpFun, Raydium, and Metaplex, enabling features like token swaps, NFT minting, and liquidity management. The framework also integrates with Langchain for natural language command interpretation and provides a Model Context Protocol (MCP) server for standardized AI agent tool invocation.

Quick Start & Requirements

• Installation: pip install agentipy
• Prerequisites: Python 3.8+, Solana CLI (for Solana actions), Langchain (pip install langchain), Wallet with private keys, Optional API keys (CoinGecko, Allora, etc.).
• Setup: Create and activate a Python virtual environment.
• Documentation: https://github.com/niceberginc/agentipy

Highlighted Details

• Extensive support for numerous Solana protocols and tools.
• Seamless integration with Langchain for natural language processing.
• Built-in MCP server for standardized AI agent interaction.
• CoinGecko integration for market data and trending tokens.

Maintenance & Community

• Community contributions are encouraged via GitHub.
• Support and updates available via email (support@agentipy.com), X (@AgentiPy), and Discord.
• Roadmap and contribution guidelines are available on GitHub.

Licensing & Compatibility

• Licensed under the MIT License.
• Permits commercial use and linking with closed-source projects.

Limitations & Caveats

• Users must securely manage private keys; examples show direct usage for demonstration only.
• Requires specific blockchain environments (Solana, Base) and associated tools (Solana CLI).