PV-Bhat/vibe-check-mcp-server

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, Ambiguous Tasks, High-Risk tasks

What it solves

Vibe Check MCP prevents AI agents from "reasoning lock-in" and over-engineering. It stops agents from confidently following flawed plans by introducing a "mentor layer" that forces them to pause and reflect before taking irreversible or complex actions.

How it works

It is a server implementing the Model Context Protocol (MCP) that acts as an external meta-cognitive signal layer. When an agent calls the vibe_check tool, the server invokes a second, independent LLM to provide feedback on the agent's current plan and goal. This implements a research-backed method called Chain-Pattern Interrupts (CPI), which injects brief pause points to re-align the agent with the user's priorities.

Who it’s for

Developers building autonomous agents who want to increase reliability, reduce harmful actions, and ensure their agents stay on the minimal viable path rather than spiraling into unnecessary complexity.

Highlights

  • CPI Adaptive Interrupts: Uses phase-aware prompts to challenge agent assumptions.
  • Multi-provider Support: Compatible with Gemini, Claude, GPT, and OpenRouter.
  • Session Constitutions: Allows users to set per-session rules (e.g., "no external network calls") that the mentor layer enforces.
  • History Continuity: Maintains context by summarizing prior advice when a session ID is provided.
  • vibe_learn: A tool to log mistakes and successes for future reflection.

Related

  • Project
    Minidoracat/mcp-feedback-enhanced

    MCP Feedback Enhanced is a security‑hardened MCP server that adds a web‑based (and optional Tauri desktop) UI for human‑in‑the‑loop feedback during long AI tasks. It works locally, over SSH, and in WSL, tracks sessions, supports images, markdown, timers, and multilingual UI, and is maintained as a fork of the original interactive‑feedback‑mcp project.

  • Project
    zhizhuodemao/js-reverse-mcp

    js‑reverse‑mcp is a Node.js MCP server that lets LLM agents debug and reverse‑engineer JavaScript in Chrome. It wraps DevTools data (scripts, breakpoints, network, WebSocket, storage) into 24 concise tools designed for AI‑driven reasoning, and includes optional stealth layers (Patchright fork and CloakBrowser) for anti‑bot sites.

  • Project
    rohitg00/agentmemory

    A persistent memory server for AI coding agents that automatically captures and retrieves project context across sessions using hybrid search.

  • Project
    spences10/mcp-sequentialthinking-tools

    An MCP server that provides a structured scratchpad for AI models to record sequential reasoning steps, validate tool plans, and manage reasoning history.

  • Project
    IBM/mcp

    A collection of Model Context Protocol (MCP) servers and tools that enable AI agents to securely interact with IBM products and enterprise resources.