functional-programming

An MCP server that exposes Crush CLI's multi-model orchestration capabilities to Claude Code and other MCP clients. Choose from 4 execution strategies based on your speed, cost, and quality needs.

```

Claude Code → Crush MCP Server → Crush CLI → Multiple AI Models

↑

(This Project)

```

---

✨ 4 Execution Strategies

• Fast: Single model, <10s, <$0.005 (grok-3-mini)
• Balanced: Two models, <30s, ~$0.015 (grok → haiku)
• Quality: Multi-model with iteration, <60s, ~$0.06 (grok → sonnet → refinement)
• Cost-Optimized: Budget-constrained, customizable max cost

🎯 MCP Tools

• crush_execute - Execute prompts with intelligent model routing
• crush_evaluate - Estimate cost/time/quality before execution

🧪 Fully Tested

• 12/12 unit tests passing
• Comprehensive strategy coverage
• TDD approach with Vitest

---

Install (2 minutes)

```bash

cd /Users/manu/Documents/LUXOR/crush-mcp-server

npm install

npm run build

npm link

```

Add to Claude Code

Already configured! ✅ Added to ~/.claude/settings.json

Next step: Restart Claude Code to load the server.

Test It

```

Use crush_execute with fast strategy to explain REST APIs in 2 sentences.

```

📖 Full guide: See [QUICKSTART.md](QUICKSTART.md)

---

Strategy Pattern

```typescript

interface Strategy {

execute(prompt: string, maxCost?: number): Promise;

}

class Orchestrator {

private strategies: Map;

async execute(request: ExecuteRequest): Promise {

const strategy = this.strategies.get(request.strategy || 'balanced');

return await strategy.execute(request.prompt, request.max_cost);

}

}

```

Execution Flow

```

┌──────────────────────────────────────────────────────────────┐

│ 1. User Prompt → Claude Code │

│ "Design a REST API using quality strategy" │

└────────────────────┬─────────────────────────────────────────┘

│ MCP Protocol

┌────────────────────▼─────────────────────────────────────────┐

│ 2. Crush MCP Server (This Project) │

│ - Parse strategy from request │

│ - Route to appropriate Strategy implementation │

└────────────────────┬─────────────────────────────────────────┘

│

┌────────────────────▼─────────────────────────────────────────┐

│ 3. Quality Strategy │

│ Step 1: grok-3-mini (outline) → Cost: $0.002 │

│ Step 2: claude-sonnet-4-5 (detail) → Cost: $0.045 │

│ Step 3: iterative refinement → Cost: $0.016 │

│ │

│ Quality Score: 0.87 (target: 0.75) │

│ Total Cost: $0.063 │

│ Execution Time: 42s │

└────────────────────┬─────────────────────────────────────────┘

│ Formatted Response

┌────────────────────▼─────────────────────────────────────────┐

│ 4. Return to User via Claude Code │

│ { │

│ result: "...", │

│ metadata: { │

│ models_used: ["grok-3-mini", "claude-sonnet-4-5"], │

│ total_cost: 0.063,