Capturing App Traffic with Stream and Parsing HAR Files Using Python

When you need to extract specific data from a mobile app or mini‑program without writing a full‑featured crawler, a lightweight approach using a proxy and a few lines of Python can automate the process.

The tutorial starts by installing the Stream app on iOS, setting up its root certificate, and enabling the capture function. All traffic from the target app (in the example, the "Jump" app for the game "The Legend of Zelda: Tears of the Kingdom") is routed through Stream.

After scrolling through the armor list in the Jump app, you stop the capture, open the capture history, and filter the requests by domain to isolate those sent by the Jump app. The filtered requests are then selected and exported as a HAR file.

On a computer, the exported HAR file is processed with a short Python script. The script uses the haralyzer library to read the HAR, extracts the text field from each response, decodes the Base64‑encoded string, and decompresses it with brotli to obtain readable JSON data.

import json
import brotli
import base64
from haralyzer import HarParser

with open('/Users/kingname/Downloads/Stream-2023-07-06 22:08:44.har') as f:
    har_parser = HarParser(json.loads(f.read()))

data = har_parser.har_data
entries = data['entries']
for entry in entries:
    text = entry['response']['content']['text']
    content = brotli.decompress(base64.b64decode(text)).decode()
    info = json.loads(content)
    print(info)

The script prints the JSON objects containing the armor data, which can then be further processed as needed. The guide notes that for most HAR files the response text is already readable, but the Jump app compresses its payload with Brotli, requiring the extra decompression step.

Finally, the article summarizes the workflow: install Stream, enable capture, browse the target app, stop capture, filter and export the HAR, and parse it with Python. Once familiar, the whole process can be completed in under three minutes.