Claude Reverse-Engineers a Gym App's API to Extract Workout Data
The author used the AI assistant Claude to reverse-engineer the API of their gym tracking app Liftoff, allowing them to export their workout data and feed it back to Claude for analysis and programming recommendations.
Why it matters
This demonstrates how AI can be used to reverse-engineer APIs and extract data from closed-off applications, empowering users to take control of their own data.
Key Points
- 1Claude used mitmproxy to intercept and analyze the Liftoff app's network traffic
- 2Claude mapped out the app's authentication flow, workout endpoint, and data structure
- 3Claude built a Go CLI tool to extract workouts, generate stats, and export data in JSON
- 4The author now uses the exported data to have Claude analyze their training and provide programming suggestions
Details
The author wanted to analyze their workout data from the Liftoff gym tracking app, but the app didn't have an export feature. They turned to the AI assistant Claude to help reverse-engineer the app's API. Claude guided the author through installing and configuring mitmproxy to intercept the app's network traffic. From the captured traffic, Claude was able to map out the app's authentication flow, workout data endpoint, and the data structure for sets and exercises. Using this information, Claude wrote a Go CLI tool that could authenticate, fetch workout data, and export it in JSON format. The author can now run this tool to get their full workout history and feed it back to Claude for analysis, allowing the AI to provide programming recommendations, spot plateaus, and flag imbalances.
No comments yet
Be the first to comment