Add Vibe Check server

[PV-Bhat]

Changes:

• Added Vibe Check Server to community server section
• Follows contribution guidelines & formatting
• Provides unique value as a novel server use case

Description

• Prevent cascading errors in AI workflows by implementing strategic pattern interrupts.
• Uses tool call "Vibe Check" with LearnLM 1.5 Pro (Gemini API), fine-tuned for pedagogy and metacognition to enhance complex workflow strategy, and prevents tunnel vision errors.
• Implements "Vibe Distill" to encourage plan simplification, prevent over-engineering solutions, and minimize contextual drift in agents.
• Self-improving feedback loops: Agents can log mistakes into "Vibe Learn" to improve semantic recall and help the oversight AI target patterns over time.

Tools:

• Vibe_check
• Vibe_distill
• Vibe_check

Motivation and Context

Vibe Check is an important tool that provides agents with self-monitoring oversight: it is a simplified way to reduce dependance on human in the loop workflows by invoking a meta-cognitive external agent, which gives the agent essential space and perspective to align more closely with user needs. Vibe Check is a first of its kind in external alignment agents implemented at the agent level.

How Has This Been Tested?

VC has been designed and tested for Claude Desktop.

Breaking Changes

Yes, MCP config JSON files will need to be changed to add the tool.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Protocol Documentation
• My changes follows MCP security best practices
• I have updated the server's README accordingly
• I have tested this with an LLM client
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have documented all environment variables and configuration options

Additional context

The server uses LearnLM 1.5 Pro (Gemini API) which is fine-tuned for guiding human learners - this is the reason I've used this model as the LLM to create the conditions for Claude (or any other AI agents) to step back and question initial assumptions without defensiveness. This has led to significant savings in API call costs, and more importantly improves the agent's alignment with user needs. It encourages ethical, moral, empathetic, and first-principles reasoning on a meta-level: forcing agents to question larger contexts even if task implementation fits guidelines.

Links:
https://smithery.ai/server/@PV-Bhat/vibe-check-mcp-server https://glama.ai/mcp/servers/@PV-Bhat/vibe-check-mcp-server https://github.com/PV-Bhat/vibe-check-mcp-server
https://mcp.so/server/vibe-check/PV-Bhat

Thank you!

[olaservo]

Thanks for your contribution to the servers list. This has been merged in this combined PR: #2007

This is a new process we're trying out, so if you see any issues feel free to re-open the PR and tag me.