store-screenshots
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill includes instructions to install the Maestro testing tool by piping a shell script from its official domain
https://get.maestro.mobile.devtobash. This is a standard installation pattern for this well-known mobile testing utility. - [EXTERNAL_DOWNLOADS]: The skill fetches documentation and configuration snippets from Fastlane's official GitHub repository via
raw.githubusercontent.com, which is a recognized and whitelisted source. - [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect injection by ingesting data from local command outputs (such as simulator lists and image metadata) to parameterize further shell commands.
- Ingestion points: Data enters the agent context via output from
xcrun simctl list,avdmanager list device, andffprobedimensions. - Boundary markers: No explicit delimiters or warnings are used to isolate these local tool outputs.
- Capability inventory: The skill has access to execute
adb,simctl,ffmpeg, andfastlaneshell commands. - Sanitization: No specific sanitization or validation is applied to the metadata gathered from local tools before it is used in command strings.
Audit Metadata