MCP ExplorerExplorer

Vibes

@jonhill90on a year ago
2 MIT
FreeCommunity
AI Systems
#ai-learning#ai-tutor#beginner-friendly#claude-desktop#conversational-ai#development-environment#docker#education#mcp#no-code
🎯 Learn anything through conversation with Claude Desktop. Give Claude superpowers with shell access, persistent memory, and code analysis. No technical background required.

Overview

What is Vibes

Vibes is an interactive learning platform that allows users to learn anything through conversation with Claude Desktop, a virtual assistant that can execute commands, analyze code, and maintain persistent memory.

Use cases

Use cases for Vibes include learning programming languages by building games, creating websites, and understanding complex concepts through guided conversations with Claude.

How to use

To use Vibes, simply set it up on your preferred platform (Mac, Windows, or Linux) following the provided guides. Once set up, you can engage in conversations with Claude to learn by asking questions and building projects together.

Key features

Key features of Vibes include shell access, persistent memory, code analysis capabilities, and a user-friendly interface designed for individuals without a technical background.

Where to use

Vibes can be used in educational settings, personal learning, software development, and any scenario where interactive learning and project building are beneficial.

Content

Ask DeepWiki

Vibes - Conversational Development Environment

Vibes transforms Claude Desktop into a conversational development environment through distributed MCP servers. Instead of learning command-line tools, you describe what you want to build and Claude implements it while teaching you.

Current State

Core Philosophy: Ask → Build → Understand → Improve → Create

Working Components:

  • Shell access and code execution in isolated containers
  • Persistent memory across conversations via vector database
  • Repository analysis and code understanding
  • INMPARA knowledge management system
  • Azure, Terraform, and GitHub integrations

Architecture: Claude Desktop + MCP servers + Docker containers + Vector DB

MCP Server Ecosystem

Server Purpose Status
mcp-vibes-server Shell access, container management ✅ Production
mcp-notebook-server INMPARA knowledge management ✅ Production
mcp-openmemory-server Persistent conversation memory ✅ Production
mcp-github-server Repository integration ✅ Production
mcp-azure-server Cloud operations ✅ Production
mcp-terraform-server Infrastructure as code ✅ Production
deepwiki-server Code analysis ✅ Production

Future Vision

Expanding toward observable AI execution and team collaboration:

Phase 1: Observable Agent Execution

  • Real-time screen sharing of AI work (terminal, browser, Neovim)
  • Agent-specific environments with persistent state
  • Pause/resume execution controls

Phase 2: Team Collaboration

  • Discord-like interface for human-AI teams
  • Multi-user knowledge spaces
  • Agent coordination visualization

Phase 3: Advanced Intelligence

  • Cross-session agent learning and skill accumulation
  • Intelligent task routing between specialized agents
  • Predictive workflow assistance

Quick Start

git clone https://github.com/jonhill90/vibes.git
cd mcp/mcp-vibes-server
docker network create vibes-network
docker-compose up -d

Configure Claude Desktop MCP settings

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "vibes": {
      "command": "docker",
      "args": [
        "exec",
        "-i",
        "mcp-vibes-server",
        "python3",
        "/workspace/server.py"
      ]
    }
  }
}

Current Capabilities

  • Execute code in safe, isolated environments
  • Remember conversations and build knowledge over time
  • Analyze any GitHub repository
  • Build real projects with infrastructure automation
  • Learn through hands-on conversation rather than documentation

Conversational development environment in production. Observable AI execution in development.

Tools

No tools

Comments

Recommend MCP Servers

View All MCP Servers