mobile-deeplink-abuse

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides detailed instructions for mobile application attacks, which presents a surface for indirect prompt injection. \n
  • Ingestion points: The body of SKILL.md. \n
  • Boundary markers: No delimiters or warning instructions are present to distinguish between instructional data and agent actions. \n
  • Capability inventory: Mentions the use of adb for interacting with mobile device services. \n
  • Sanitization: No sanitization logic for parameters in the provided command templates. \n- [COMMAND_EXECUTION]: Includes functional adb command-line templates for triggering intents and activities on mobile devices, which could be misused if executed by an agent without oversight. \n- [NO_CODE]: This skill consists of documentation and metadata only, with no executable scripts or code files provided in the package.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 11:02 PM
Security Audit — agent-trust-hub — mobile-deeplink-abuse