Installation

FrootAI is available through multiple distribution channels. Choose the one that fits your workflow.

Prerequisites

Requirement	Minimum Version	Check
Node.js	18+ (22+ recommended)	`node --version`
Python	3.9+ (for Python SDK)	`python --version`
Docker	20+ (for container usage)	`docker --version`
Git	2.30+	`git --version`
Azure CLI	2.60+ (for deployment)	`az version`

ℹ️

Only Node.js is required for the core tooling. Python, Docker, and Azure CLI are needed only for their respective channels.

Quick Install

Choose your preferred channel:

npm


# MCP Server — add AI tooling to any MCP-compatible agent
npx frootai-mcp@latest
 
# SDK — programmatic API for Node.js / TypeScript
npm install frootai
 
# CLI — command-line access
npm install -g frootai

pip


# Python SDK — offline, zero dependencies
pip install frootai
 
# Python MCP Server
pip install frootai-mcp

Docker


# Zero-install MCP server
docker pull ghcr.io/frootai/frootai-mcp
docker run -i ghcr.io/frootai/frootai-mcp

VS Code


# Install the VS Code extension
code --install-extension frootai.frootai-vscode

Or search “FrootAI” in the Extensions panel (Ctrl+Shift+X).

MCP Server Configuration

Add FrootAI as an MCP server in your editor:

VS Code / Copilot

.vscode/mcp.json


{
  "mcpServers": {
    "frootai": {
      "command": "npx",
      "args": ["frootai-mcp@latest"]
    }
  }
}

Claude Desktop

claude_desktop_config.json


{
  "mcpServers": {
    "frootai": {
      "command": "npx",
      "args": ["frootai-mcp@latest"]
    }
  }
}

Cursor

.cursor/mcp.json


{
  "mcpServers": {
    "frootai": {
      "command": "npx",
      "args": ["frootai-mcp@latest"]
    }
  }
}

Docker Compose

docker-compose.yml


services:
  frootai-mcp:
    image: ghcr.io/frootai/frootai-mcp:latest
    stdin_open: true

💡

The MCP Server exposes 45 tools to any AI agent that speaks the Model Context Protocol .

SDK Usage

Use FrootAI programmatically in your applications:

JavaScript


import { FrootAI } from 'frootai';
 
const fai = new FrootAI();
fai.search('RAG architecture');   // BM25-ranked results
fai.plays.get('01');              // Play details
fai.listModules();                // 16 FROOT modules

Python


from frootai import FrootAI
 
fai = FrootAI()
results = fai.search("RAG architecture")
play = fai.plays.get("01")
modules = fai.list_modules()

CLI


# Search across all knowledge
npx frootai search "RAG architecture"
 
# Get play details
npx frootai plays get 01
 
# List modules
npx frootai modules list

Clone the Repository

For contributing or exploring the full source:


git clone https://github.com/frootai/frootai.git
cd frootai
npm install
npm run validate:primitives  # Verify everything works

⚠️

The repository includes 100 solution plays, 238 agents, 322 skills, and 77 plugins. The initial clone is approximately 50 MB.

Distribution Channels Summary

Channel	Install Command	Use Case
npm MCP	`npx frootai-mcp@latest`	Add FrootAI tools to any AI agent
npm SDK	`npm install frootai`	Programmatic API in Node.js/TS
PyPI SDK	`pip install frootai`	Python API, offline, zero deps
PyPI MCP	`pip install frootai-mcp`	Python MCP server
VS Code	`code --install-extension frootai.frootai-vscode`	IDE integration
Docker	`docker run -i ghcr.io/frootai/frootai-mcp`	Zero-install MCP server
CLI	`npx frootai`	Command-line access
GitHub	`git clone`	Full source, contributions

Next Steps

Quick Start — 5-minute guide from zero to running
Your First Solution Play — build and deploy Play 01