Problem Statement Distraction rules are currently static — you have to manually anticipate every app and website that might pull you off track. In practice, distractions are contextual: Reddit might be a distraction during a coding session but legitimate research during a content project. And new distractions pop up constantly — you only realize you should have blocked something after you've already lost 30 minutes to it. Meanwhile, the AI Insights feature already has access to all tracking data but only works retrospectively. It can tell you what happened, but it can't intervene while it's happening. Proposed Feature: AI-Powered Distraction Detection Use the existing AI layer to monitor recent activity in the context of your current project and proactively detect distractions: AI compares your current activity (app, website, window title) against your active project's category, description, and allowed apps When it detects something that looks off-track, send a native notification: "You've been on YouTube for 15 min — doesn't match your AWS Study project. Distraction?" The notification offers a one-click action: "Add to blocked list" which creates a new distraction rule automatically Over time, the AI learns your patterns and suggests new distraction rules you can accept or dismiss Why It Fits Chronoid The building blocks are already there — automatic tracking, distraction rules, AI with data access, and the web blocker. This feature connects them into a closed loop: detect, notify, block. Instead of users manually maintaining a rule list, the AI keeps it current based on actual behavior. This would work well with both local models (Ollama) and cloud providers, since the context needed is minimal — just recent activity + active project metadata. Benefit Turns distraction management from a manual setup task into an adaptive system that gets smarter as you use it. You stop needing to predict distractions in advance — Chronoid catches them in context.
