Vibe Check
Vibe Check is a tool that provides mentor-like feedback to AI Agents, preventing tunnel-vision, over-engineering and reasoning lock-in for complex and long-horizon agent workflows. KISS your over-eager AI Agents goodbye! Effective for: Coding, Ambigu
What is Vibe Check?
Vibe Check is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to vibe check is a tool that provides mentor-like feedback to ai agents, preventing tunnel-vision, over-engineering and reasoning lock-in for complex and long-horizon agent workflows. kiss your over-eage...
Vibe Check is a tool that provides mentor-like feedback to AI Agents, preventing tunnel-vision, over-engineering and reasoning lock-in for complex and long-horizon agent workflows. KISS your over-eager AI Agents goodbye! Effective for: Coding, Ambigu
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Vibe Check is a tool that provides mentor-like feedback to A
Use Cases
Maintainer
Works with
Installation
NPM
npx -y @pv-bhat/vibe-check-mcpManual Installation
npx -y @pv-bhat/vibe-check-mcpConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Vibe Check
Vibe Check MCP is a metacognitive feedback server that acts as a second-opinion mentor for AI agents engaged in complex, long-horizon tasks. It exposes a vibe_check tool that invokes a separate LLM to evaluate the agent's current plan and strategy, surfacing risk scores, uncertainty levels, and potential blind spots before the agent commits to costly or irreversible actions. Additional tools (vibe_learn, update_constitution, reset_constitution, check_constitution) let agents capture lessons from past mistakes and maintain session-specific behavioral rules. Developers running autonomous coding, research, or decision-making agents will find Vibe Check especially useful for preventing tunnel vision and over-engineering.
Prerequisites
- Node.js 18 or later installed
- An API key for at least one supported LLM provider: Gemini (GEMINI_API_KEY, default), OpenAI (OPENAI_API_KEY), OpenRouter (OPENROUTER_API_KEY), or Anthropic (ANTHROPIC_API_KEY)
- An MCP-compatible client such as Claude Desktop or Cursor
Choose your LLM provider and get an API key
Vibe Check uses a second LLM to review the primary agent's plan. Gemini is the default provider. Get an API key from Google AI Studio (Gemini), OpenAI, OpenRouter, or Anthropic.
Add the server to your MCP client configuration
Open your claude_desktop_config.json (or equivalent) and add the vibe-check-mcp entry with your chosen LLM provider's API key.
{
"mcpServers": {
"vibe-check": {
"command": "npx",
"args": ["-y", "@pv-bhat/vibe-check-mcp", "start", "--stdio"],
"env": {
"GEMINI_API_KEY": "your_gemini_api_key",
"DEFAULT_LLM_PROVIDER": "gemini"
}
}
}
}Optionally switch to OpenAI or OpenRouter
If you prefer a different provider, set the appropriate env var and update DEFAULT_LLM_PROVIDER. You can also specify a DEFAULT_MODEL.
{
"mcpServers": {
"vibe-check": {
"command": "npx",
"args": ["-y", "@pv-bhat/vibe-check-mcp", "start", "--stdio"],
"env": {
"OPENAI_API_KEY": "your_openai_key",
"DEFAULT_LLM_PROVIDER": "openai",
"DEFAULT_MODEL": "gpt-4o"
}
}
}
}Restart your MCP client
Save the config and restart Claude Desktop or your chosen client to load the vibe-check server.
Instruct your agent to call vibe_check before major actions
Add a system prompt or instruction telling the agent to invoke vibe_check after planning a solution and before executing significant, hard-to-reverse steps.
# Example agent system prompt addition:
# Before taking any major action (file deletion, API calls, large rewrites),
# call vibe_check with the full user request and your current plan.
# Proceed only if vibe_check confirms low risk.Vibe Check Examples
Client configuration
Claude Desktop config for Vibe Check MCP using Gemini as the reviewer LLM.
{
"mcpServers": {
"vibe-check": {
"command": "npx",
"args": ["-y", "@pv-bhat/vibe-check-mcp", "start", "--stdio"],
"env": {
"GEMINI_API_KEY": "your_gemini_api_key",
"DEFAULT_LLM_PROVIDER": "gemini"
}
}
}
}Prompts to try
Example invocations of the Vibe Check tools during agent workflows.
- "Before refactoring this entire module, run a vibe_check on your plan"
- "Call vibe_check with the user's request and your proposed solution before writing any code"
- "After fixing that bug, use vibe_learn to record what went wrong and how you resolved it"
- "Check the current session constitution before proceeding"
- "Reset the constitution and start fresh for this new task"Troubleshooting Vibe Check
vibe_check returns an error about missing API key
Ensure the environment variable for your chosen provider is set in the MCP config env block. The default provider is Gemini — set GEMINI_API_KEY unless you have also set DEFAULT_LLM_PROVIDER to another value.
Server starts but vibe_check is very slow or times out
The tool makes a live LLM API call, so response time depends on the provider. Switch to a faster model by setting DEFAULT_MODEL (e.g., 'gemini-1.5-flash' for Gemini) to reduce latency.
npx fails to download or install the package
Run 'npx -y @pv-bhat/vibe-check-mcp start --stdio' manually in a terminal to see the full error. Ensure npm registry access and that Node.js 18+ is installed. Clear the npx cache if a stale version is cached.
Frequently Asked Questions about Vibe Check
What is Vibe Check?
Vibe Check is a Model Context Protocol (MCP) server that vibe check is a tool that provides mentor-like feedback to ai agents, preventing tunnel-vision, over-engineering and reasoning lock-in for complex and long-horizon agent workflows. kiss your over-eager ai agents goodbye! effective for: coding, ambigu It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Vibe Check?
Install via npm with the command: npx -y @pv-bhat/vibe-check-mcp. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with Vibe Check?
Vibe Check works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Vibe Check free to use?
Yes, Vibe Check is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
Vibe Check Alternatives — Similar Coding Agents Servers
Looking for alternatives to Vibe Check? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Vibe Check in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Vibe Check?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.