activity-analyzer
by qew21
CLI UtilitiesBrief
Use ActivityWatch to analyze user's computer activity (Requires Node.js)
Description
What This Skill Does
Connects to a locally running ActivityWatch instance to fetch computer activity data and run an AI-powered productivity analysis. The agent summarizes time spent per application, flags distracting patterns, and gives specific behavioral suggestions. Window titles are included by default, so data containing sensitive content is sent to the model for processing.
ActivityWatch stores data locally on 127.0.0.1, so no third-party time-tracking account or cloud sync is required before analysis.
When to Use It
- Reviewing how much coding time vs meeting time was logged today
- Spotting excessive YouTube or Reddit usage during work hours
- Understanding app-switching frequency on high-distraction days
- Getting a structured daily screen-time breakdown without a paid tracker
- Auditing whether deep-work blocks actually happened as planned
Example Workflow
User asks: Give me a summary of my computer activity for the last 24 hours
- 1
Runs
node scripts/fetch_activity.js --hours 24against the local ActivityWatch API - 2
Checks whether output contains raw window titles and warns the user if sensitive content is detected
- 3
Groups activity by application and calculates time spent in each
- 4
Identifies anomalies such as frequent context switches or extended use of non-work sites
- 5
Produces a time-distribution summary and 2-3 specific productivity suggestions
A breakdown of time per app category, flagged distractions (e.g., 3h 20m on YouTube), and concrete advice such as scheduling focused blocks using the Pomodoro technique
Requirements
- ActivityWatch installed and running locally (activitywatch.net)
Installation
clawhub install qew21/activity-analyzerRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.