mobile-automation

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from mobile application interfaces which could contain adversarial instructions targeting the agent's logic.
  • Ingestion points: mobile_list_elements_on_screen and mobile_take_screenshot in SKILL.md ingest UI text and visual data from external apps.
  • Boundary markers: Absent. The instructions do not specify delimiters or warnings to ignore embedded content within the mobile UI.
  • Capability inventory: High-impact capabilities include mobile_click_on_screen_at_coordinates, mobile_type_keys, mobile_open_url, and mobile_login_to_cloud_provider in SKILL.md.
  • Sanitization: Absent. There is no evidence of sanitization or filtering of screen content before the agent interprets it.
  • [COMMAND_EXECUTION]: The skill metadata specifies a requirement for npx, indicating it relies on shell-level command execution to manage its operational environment.
  • [EXTERNAL_DOWNLOADS]: The skill instructions reference documentation hosted at mobilenext.ai, which is the official domain for the skill's author, mobile-next.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 08:13 AM
Security Audit — agent-trust-hub — mobile-automation