Tinyman MCP
An MCP server for the Tinyman protocol on the Algorand blockchain, offering tools for swaps, liquidity provision, and pool management.
Tinyman MCP
A Model Context Protocol (MCP) implementation for Tinyman on Algorand, built by GoPlausible.
Overview
This project provides MCP tools and resources for interacting with Tinyman AMM protocol on Algorand blockchain. It consists of two main packages:
@tinyman-mcp/server
: MCP server implementation providing tools for Tinyman operations like swaps, liquidity provision, pool management and analytics@tinyman-mcp/client
: Client implementation for wallet connections and transaction signing (Work in Progress)
Features
The MCP server provides the following tools:
Pool Management
tinyman_get_pool
: Retrieve detailed information about a specific pooltinyman_get_pool_analytics
: Get analytics and statistics for a pooltinyman_get_pool_creation_quote
: Generate quote for creating a new pool
Trading Operations
tinyman_get_swap_quote
: Generate quotes for asset swaps with support for:- Fixed input amount swaps
- Fixed output amount swaps
- Slippage calculations
Liquidity Management
tinyman_get_liquidity_quote
: Generate quotes for adding liquidity- Supports single and dual asset deposits
- Calculates expected pool tokens
tinyman_get_remove_liquidity_quote
: Generate quotes for removing liquidity- Supports proportional and single-asset withdrawals
- Calculates expected asset returns
Asset Operations
tinyman_get_asset_optin_quote
: Generate quote for opting into an assettinyman_get_validator_optin_quote
: Generate quote for opting into a validatortinyman_get_validator_optout_quote
: Generate quote for opting out of a validator
All tools support both Tinyman v1.1 and v2 protocols, with v2 being the default version where applicable.
Installation
To install or update the Tinyman MCP implementation, clone the repository, install the dependencies and build the project":
First check node version to be 23.6.1 or later:
node -v
Upgrade to 23.6.1 or later if needed!
Then check the Claude or Cursor container folders to have mcp-servers folder (if not create one):
mkdir PATH_ON_YOUR_MACHINE/Claude/mcp-servers
# or for Cursor
mkdir PATH_ON_YOUR_MACHINE/Cursor/mcp-servers
Then clone this repository under mcp-servers folder and install dependencies:
cd PATH_ON_YOUR_MACHINE/Claude/mcp-servers
# or for Cursor
cd PATH_ON_YOUR_MACHINE/Cursor/mcp-servers
# Clone the repository
git clone https://github.com/GoPlausible/tinyman-mcp.git
cd tinyman-mcp
# Install dependencies
npm install
# Build the project
npm run build
# Edit the .env file to set your configurations
And you are done! Now you can open you MCP config and add the server as :
{
"mcpServers": {
"tinyman-mcp": {
"command": "node",
"args": [
"PATH_ON_YOUR_MACHINE/Claude/mcp-servers/tinyman-mcp/packages/server/dist/index.js"
],
"env": {
"ALGORAND_NETWORK": "mainnet",
"ALGORAND_ALGOD_API": "https://mainnet-api.algonode.cloud/v2",
"ALGORAND_ALGOD": "https://mainnet-api.algonode.cloud",
"ALGORAND_ALGOD_PORT": "",
"ALGORAND_TOKEN": "",
"ALGORAND_AGENT_WALLET": "problem aim online jaguar upper oil flight stumble mystery aerobic toy avoid file tomato moment exclude witness guard lab opera crunch noodle dune abandon broccoli",
"ITEMS_PER_PAGE": "10"
}
}
}
}
Make sure yopu change the paths to match your local system's paths.
For example on MACOS and Claud, the path would be something like this:
{
"mcpServers": {
"algorand-mcp": {
"command": "node",
"args": [
" /Users/YOUR_USERNAME/Library/Application\ Support/Claude/mcp-servers/tinyman-mcp/packages/server/dist/index.js"
]
}
}
}
Project Structure
packages/
├── client/ # Client implementation (WIP)
└── server/ # MCP server implementation
Development
# Install dependencies
npm install
# Build all packages
npm run build
# Clean build artifacts
npm run clean
License
MIT
Related Servers
MCP Quickstart
A basic MCP server from the Quickstart Guide, adapted for OpenAI's Chat Completions API.
Replicate Ideogram V3 Balanced
Generate images using the Ideogram V3 Balanced model on Replicate.
MCPunk
Explore and understand codebases through conversation by breaking files into logical chunks for searching and querying without embeddings.
Gemini MCP Server
An MCP server for interacting with Google's Gemini models through the Gemini CLI.
ITerm MCP Server
An MCP server for iTerm2, enabling AI assistants to interact with the terminal.
Authenticator App MCP Server
A secure MCP server for AI agents to interact with the Authenticator App for 2FA codes and passwords.
CodeBase Optimizer
Analyzes, optimizes, and detects duplicates in codebases for Claude Code.
Sui MCP Tools
A toolkit for interacting with the Sui blockchain and integrating MCP SDK features, with support for multiple network environments.
Sensei MCP
Expert guidance for Dojo and Cairo development on Starknet, specializing in the Dojo ECS framework for building onchain worlds.
MCP Talk Demo Files
A collection of demo files for MCP servers and clients, illustrating various transport protocols and server capabilities using Python.