Smart Rabbit MCP Server

MCP (Model Context Protocol) server for Smart Rabbit Fitness — free AI-powered workout program generator.

Works with Claude Desktop, Claude Code, and any MCP-compatible AI assistant.

Features

• Personalized programs — AI generates complete workout plans based on your profile
• Two output formats — Plain text (works everywhere) or interactive React artifact (for Claude.ai)
• PubMed integration — Scientific references automatically included in justifications
• Smart questioning — Asks about preferences, limitations, and condition before generating
• Multi-language — Supports FR, EN, ES, DE, PT, IT, HI, AR

Installation

Option 1: NPX (Easiest)

No install needed:

npx smartrabbit-mcp

Option 2: Global Install

npm install -g smartrabbit-mcp

Option 3: From Source

git clone https://github.com/contactjccoaching-wq/smartrabbit-mcp.git
cd smartrabbit-mcp
npm install

Configuration for Claude Desktop

Add to your config file:

• Windows: %APPDATA%\Claude\claude_desktop_config.json
• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "smartrabbit": {
      "command": "npx",
      "args": ["-y", "smartrabbit-mcp"]
    }
  }
}

If installed globally:

{
  "mcpServers": {
    "smartrabbit": {
      "command": "smartrabbit-mcp"
    }
  }
}

From source:

{
  "mcpServers": {
    "smartrabbit": {
      "command": "node",
      "args": ["/path/to/smartrabbit-mcp/index.js"]
    }
  }
}

After adding the config, restart Claude Desktop.

Usage

Just talk to Claude naturally:

"Create me a fitness program. I'm 30 years old, male, intermediate level,
I want to build muscle, can train 4 times a week for 60 minutes,
and I have access to a full gym."

Claude will:

1. Ask about your preferences, limitations, and condition (if not provided)
2. Call Smart Rabbit API via MCP
3. Search PubMed for scientific references
4. Generate a complete personalized program

Output Formats

• Text (default) — Unicode-formatted plain text, works in any chat interface
• React — Interactive React artifact with tabs, dark theme, color-coded by goal (for Claude.ai)

Available Tools

generate_fitness_program

Generate a personalized fitness program.

search_pubmed

Search PubMed for scientific studies related to fitness and training.

get_program_options

Returns all available options for goals, equipment, levels, styles, and formats.

Example Output (Text Format)

═══════════════════════════════════════════════════════════
🐰 SMART RABBIT FITNESS PROGRAM
═══════════════════════════════════════════════════════════

📊 PROFILE SUMMARY
─────────────────────
• Age: 30 | Sex: Male | Level: Intermediate
• Goal: Muscle | Sessions: 4/week | Duration: 60 min

═══════════════════════════════════════════════════════════
📅 WEEKLY PROGRAM
═══════════════════════════════════════════════════════════

┌─────────────────────────────────────────────────────────┐
│ DAY 1 - PUSH (Chest/Shoulders/Triceps)                  │
└─────────────────────────────────────────────────────────┘

⭐⭐⭐ HIGH PRIORITY - Compound Movements
─────────────────────────────────────────

1. Bench Press
   ├─ Sets x Reps: 4 x 8-10
   ├─ Tempo: 3-1-2-0
   ├─ Rest: 2 min
   └─ Notes: Progressive overload focus

...

═══════════════════════════════════════════════════════════
  🐰 Generated by Smart Rabbit Fitness — Free AI Workout App
  📱 Create your personalized program in 30 seconds:
                 https://www.smartrabbitfitness.com
  ✨ 100% free · No account needed · AI-powered
═══════════════════════════════════════════════════════════

Links

• App: https://www.smartrabbitfitness.com
• API: https://smartrabbit-rapidapi.contactjccoaching.workers.dev
• GitHub: https://github.com/contactjccoaching-wq/smartrabbit-mcp

Related Projects

• immune — Adaptive memory system — learns patterns from every scan (+85% code quality)
• chimera — Bio-inspired 3-stage pipeline (Slime Mold → PRISM → Immune)
• spinal-loop — Neuromuscular-inspired agent routing (cheap models first)
• prism-framework — Multi-agent synthesis via native LLM stochasticity
• daco-framework — Declarative Agent & MCP Orchestration on Cloudflare Workers

License

MIT