skills/zfinix/another/another-android/Gen Agent Trust Hub

another-android

Warn

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The another_shell tool allows for the execution of arbitrary adb shell commands on the target Android device. This provides full access to the device's shell environment, enabling tasks such as file system modification, process management, and access to system logs.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a significant attack surface for indirect prompt injection due to its ability to ingest untrusted data and perform high-privilege actions.
  • Ingestion points: The agent receives external, potentially attacker-controlled data via another_take_screenshot (visual content) and another_shell (command output).
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands found within the device's UI or shell output.
  • Capability inventory: The agent has access to powerful tools including arbitrary shell execution (another_shell), URL navigation (another_open_url), and simulated user input (another_send_text, another_send_touch).
  • Sanitization: There is no evidence of command sanitization or input validation for data retrieved from the device before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 26, 2026, 12:21 AM