🧠 Vibe Check MCP - Overview

Title

Vibe Check MCP – Your AI's inner rubber duck when it can't rubber duck itself.

How to Use

Vibe Check MCP integrates with AI agents (like Claude) to provide metacognitive oversight, preventing over-engineering and misaligned solutions.

Installation

• Via Smithery (Auto-install for Claude Desktop):

  npx -y @smithery/cli install @PV-Bhat/vibe-check-mcp-server --client claude  

• Manual Setup (npm):

  git clone https://github.com/PV-Bhat/vibe-check-mcp-server.git  
  cd vibe-check-mcp-server  
  npm install  
  npm run build  
  npm run start  

Integration

Add to your claude_desktop_config.json:

"vibe-check": {  
  "command": "node",  
  "args": ["/path/to/vibe-check-mcp/build/index.js"],  
  "env": { "GEMINI_API_KEY": "YOUR_GEMINI_API_KEY" }  
}  

Environment Setup

Create a .env file with:

GEMINI_API_KEY=your_gemini_api_key_here  

Key Features

1. 🛑 vibe_check – Pattern interrupt to break tunnel vision in AI agents.
2. ⚓ vibe_distill – Simplifies complex plans to align with user intent.
3. 🔄 vibe_learn – Self-improving feedback loop to refine future responses.
4. Metacognitive Oversight – Prevents over-engineering, scope creep, and misalignment.
5. Phase-Specific Interrupts – Works during planning, implementation, and review phases.

Use Cases

• Preventing Over-Engineering – Stops AI from turning a simple CSV parser into a full-scale enterprise solution.
• Correcting Misalignment – Ensures AI solves the actual problem, not a misinterpreted one.
• Simplifying Workflows – Forces agents to distill complex plans into essential steps.
• Self-Improving Agents – Logs mistakes via vibe_learn to enhance future responses.

When to Use Each Tool

For full details, check the Technical Reference.

Links:
GitHub | Smithery | mcp.so