ContextPulse/contextpulse

What it does

ContextPulse is a desktop context server designed for AI agents that captures and exposes comprehensive contextual information about your computer environment. It functions as an MCP (Model Context Protocol) server, enabling AI assistants to access real-time data about what's happening on your screen, what you're saying, and your input patterns.

Key features

• Screen capture with OCR (optical character recognition) for text extraction from visual content
• Voice transcription powered by OpenAI Whisper
• Keyboard and mouse activity tracking
• Clipboard monitoring
• Semantic memory storage and retrieval
• Automatic project detection
• Activity history with temporal context
• 35+ MCP tools for various context operations
• Zero-cloud local-first architecture for privacy
• Cross-platform support (macOS, Windows, Linux)

How to set up

Clone the repository from GitHub, install Python dependencies, configure the MCP server settings, and connect it to your preferred AI platform (Claude, ChatGPT, or other MCP-compatible clients). The local-first design means all processing occurs on your machine with no data sent to external servers.