We all know the drill: you want to build a habit, but updating your tracker feels like a chore. Enter TrackTogether. TrackTogether is a MCP native accountability platform designed from the ground up to let your AI agents (Claude, ChatGPT, or any MCP compatible tool) read, update, and analyze your progress directly. No more context switching. While you work alongside your AI assistant, it seamlessly logs your habits in the background.
I know what you're thinking: the job search tracker space is completely saturated. But the real friction isn't the tracking it's the context switching. Since most of us already live inside Claude or ChatGPT all day to review resumes, debug code, and do mock interviews etc, I realized the tracker should live there too.
To solve this, I built a remote Model Context Protocol (MCP) server with native OAuth. Now, instead of opening a separate app, I just tell my AI assistant directly to log a solved problem, update an interview stage, or pull up my pipeline.
Would love any feedback on the concept, especially from anyone else building MCP integrations. If anyone is wrestling with setting up remote auth flows over MCP, I’m happy to talk through how I structured mine!