DumbMoney MCP Server
# DumbMoney MCP Server
MCP (Model Context Protocol) server for [DumbMoney](https://dumbmoney.win) — query reflection tokens, check earnings, and discover top earners on Solana.
## Tools
| Tool | Description |
|------|-------------|
| `list_tokens` | List all reflection tokens with price, market cap, and bonding curve progress |
| `get_token` | Get detailed info about a specific token by mint address |
| `check_earnings` | Check a wallet's pending reflection earnings for a token |
| `top_earners` | Get the top 10 tokens by total reflections paid |
## Usage with Claude Code
Add to your `~/.claude/settings.json`:
```json
{
"mcpServers": {
"dumbmoney": {
"command": "npx",
"args": ["@dumbmoney/mcp-server"]
}
}
}
```
## Usage with Claude Desktop
Add to your Claude Desktop config:
```json
{
"mcpServers": {
"dumbmoney": {
"command": "npx",
"args": ["@dumbmoney/mcp-server"]
}
}
}
```
## Build from source
```bash
cd mcp-server
npm install
npm run build
npm start
```
TDQS
Scored across 7 tools
Each tool targets a distinct resource and action: list_tokens provides an overview, get_token drills into a specific mint, check_earnings focuses on a wallet's pending rewards, top_earners is a ranking, create_token launches a new token, register_agent handles account creation, and get_my_agent_info shows the agent's profile. No two tools have overlapping primary purposes.
The vast majority follow a clear verb_noun snake_case pattern (list_tokens, get_token, check_earnings, create_token, register_agent, get_my_agent_info). The exception is top_earners, which uses an adjective_noun form instead of a verb_noun form, causing a minor inconsistency.
With 7 tools, the server is well-scoped for its purpose. It provides enough functionality to cover token discovery, earnings tracking, token creation, and agent management, without unnecessary bloat or gaps.
The tool surface covers the core workflows: listing and inspecting tokens, checking earnings, creating tokens, and managing agent credentials. Minor gaps exist, such as no tool to list tokens created by the agent or a direct action to claim pending reflections, but these do not severely block main use cases.