The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/check_environment.py utilizes subprocess.run to execute system commands including adb, emulator, and avdmanager to verify the Android development environment. While these are currently limited to status checks, they represent an execution surface for local tools.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by ingesting untrusted content from APK files during the analysis phase.
Ingestion points: scripts/analyze_apk.py extracts activity names, permissions, and string resources from user-provided APK files.
Boundary markers: Absent. The extracted metadata is directly used in the workflow to generate user-facing test cases and scenarios.
Capability inventory: The skill possesses the ability to execute shell commands via adb and perform network operations via urllib and appium client.
Sanitization: Absent. No validation or escaping is performed on the extracted APK strings before they are presented to the agent for logic generation.
[EXTERNAL_DOWNLOADS]: The skill performs a network request to http://localhost:4723/status using urllib.request.urlopen to check the status of the local Appium server. While targeting localhost, this involves processing external JSON data from a network service.
[COMMAND_EXECUTION]: In scripts/check_environment.py, the skill uses the __import__ function to dynamically load Python modules (pytest, jinja2, pillow) to verify that the required environment dependencies are installed.

test-mobile-app