Skill Detail

ActivityWatch Cross-Platform Automated Time Tracker and Productivity Analyzer

ActivityWatch is a privacy-first, open-source automated time tracker that records application usage, browser activity, and AFK status across Windows, macOS, and Linux. With 16k+ GitHub stars, it provides detailed productivity analytics without sending data to external servers.

Calendar, Email & ProductivityMulti-Framework
Calendar, Email & Productivity Multi-Framework Security Reviewed
โญ 17.2k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill activitywatch-automated-time-tracker Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Author
ActivityWatch
Last updated
Apr 4, 2026
Quick brief

ActivityWatch is a cross-platform automated time tracking application designed to collect valuable lifedata while keeping user privacy at the forefront. With over 16,000 GitHub stars and an active development community, it has become the leading open-source alternative to services like RescueTime and Toggl Track.

How it works

What this skill actually does

How It Works

ActivityWatch runs a local server (aw-server or aw-server-rust) that collects events from watchers โ€” lightweight background processes that monitor specific activity sources. The two core watchers included are aw-watcher-window (tracks the currently active application and window title) and aw-watcher-afk (detects keyboard and mouse idle states). A browser extension (aw-watcher-web) tracks active tabs and URLs in Chrome and Firefox.

Key Features

All data stays on your local machine. The web-based dashboard provides timeline views, activity summaries, and category breakdowns. A built-in query language lets you write custom reports and filter events by application, window title, URL, or time range. The REST API on localhost:5600 enables programmatic access to all collected data for custom integrations and exports.

Integration Points for AI Agents

The localhost REST API makes ActivityWatch particularly useful for AI agent workflows. Agents can query productivity data to generate daily summaries, detect focus patterns, identify time sinks, or trigger alerts when certain thresholds are reached. The JSON export format is straightforward to parse and transform. Community watchers extend tracking to VS Code editor activity, Spotify playback, and custom data sources.

Technical Stack

The server is available in Python (aw-server) or Rust (aw-server-rust) implementations. The web UI uses Vue.js. Desktop builds are distributed as standalone bundles for Windows, macOS, and Linux. An Android app provides mobile tracking. The project uses the Mozilla Public License 2.0, allowing both open-source and proprietary integrations.