tailrocks-macos-visual-regression
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
bun,osascript,defaults, andxcodebuildto automate visual regression workflows, including application launching, window capture, and accessibility tree traversal. - [PRIVILEGE_ESCALATION]: The skill modifies system-wide appearance and accessibility settings (e.g.,
increaseContrast,ReduceTransparency, and Dark Mode) viadefaults writeandosascript. These actions are necessary to establish a deterministic testing environment and are explicitly disclosed as core functionality with built-in restoration and verification steps. - [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by processing external baseline configuration files and application metadata (accessibility identifiers) while possessing the capability to execute system-level commands.
- Ingestion points: Processes
BASELINE.jsonmanifests, application window attributes, and render captures. - Boundary markers: Employs strict schema validation, identity binding, and digest verification for all baseline rows and artifacts to ensure data integrity.
- Capability inventory: Includes system setting modification and shell command execution via local scripts and system tools.
- Sanitization: Uses explicit digest checks and schema enforcement to prevent the processing of malformed or unauthorized baseline data.
Audit Metadata