About savvyscratch
savvyscratch is an MCP server that lets AI assistants access scratch-off lottery data and analysis for 19 U.S. states through natural language. It also supports account registration, subscription management, and budget-based game selection.
Key features
- 01Account registration, login, and subscriptions
- 02Browse scratch-off games by state
- 03View detailed game prize breakdowns
- 04Find games by budget
- 05Free previews and subscriber access
Use cases
- 01Query scratch-off games in a state
- 02Compare games within a budget
- 03Create an account or subscribe through an AI assistant
MCP support
It provides account and game-analysis tools to compatible clients as a stdio MCP server.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
9Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| string | Yes | Your email address | |
| password | string | Yes | Choose a password (minimum 6 characters) |
Details
From the project READMESavvy Scratch MCP Server
Install
Claude Desktop
Open your claude_desktop_config.json:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add the Savvy Scratch server:
{
"mcpServers": {
"savvyscratch": {
"command": "npx",
"args": ["-y", "@savvyscratch/mcp-server"]
}
}
}
Restart Claude Desktop. You're done.
Claude Code
Add to your project's .mcp.json (or run claude mcp add):
{
"mcpServers": {
"savvyscratch": {
"command": "npx",
"args": ["-y", "@savvyscratch/mcp-server"]
}
}
}
Other MCP-compatible clients
Any client that supports stdio MCP servers can run:
npx -y @savvyscratch/mcp-server
What is this?
An MCP (Model Context Protocol) server that gives AI assistants access to Savvy Scratch — pro-gambler-grade analysis of scratch-off lottery games across 19 U.S. states.
Register, sign in, browse live game data, and subscribe — all through natural conversation with your AI. Zero configuration required.
First-time setup (through your AI)
Once installed, just talk to your AI:
- "Register a Savvy Scratch account for me" → AI uses the
registertool - "What scratch-off games should I play in Florida?" → AI uses
login+get_games - "Subscribe me to Savvy Scratch" → AI uses
get_subscribe_link
No API keys. No env vars. Your account lives in your AI session.
Tools
Auth & Account
| Tool | Description |
|---|---|
register | Create a new Savvy Scratch account |
login | Sign in with email and password |
logout | Sign out |
check_subscription | View subscription status and plan details |
get_subscribe_link | Get a link to subscribe |
Game Analysis
| Tool | Description | Free | Subscriber |
|---|---|---|---|
list_states | List all 19 covered states | Full | Full |
get_games | Get scratch-off games for a state | Top 3 | All games |
game_details | Detailed prize breakdown for a game | Locked | Full |
best_games | Best games for a budget and state | Top 3 | All matches |
Pricing
Free accounts see top-3 previews in every state. A Savvy Scratch subscription ($5/mo or $50/yr) unlocks:
- All games in all 19 states
- Detailed prize-tier breakdowns with live remaining prizes
- Budget-optimized recommendations ranked by weighted score
- Pro gambler methodology applied to every game
Example Conversation
You: What scratch-off games should I play in Florida?
AI: [calls login, then get_games with state=fl]
Shows top 3 games with scores and odds...
"Subscribe to see all 100 games. Use get_subscribe_link to get started."
You: I have $10 to spend in Georgia, what's best?
AI: [calls best_games with state=ga, budget=10]
Shows top-rated games at $10 or under ranked by weighted score
You: Show me the prize breakdown for Atlanta Falcons
AI: [calls game_details with state=ga, game_name="Atlanta Falcons"]
Full prize tier breakdown with remaining prizes and adjusted odds
Build from source
git clone https://github.com/myque08/savvymcp.git
cd savvymcp
npm install
npm run build
node dist/index.js
Links
- npm package: https://www.npmjs.com/package/@savvyscratch/mcp-server
- Website: https://www.savvyscratch.com
- GitHub: https://github.com/myque08/savvymcp
- Report issues: https://github.com/myque08/savvymcp/issues
Comments
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.