sdr-rf-attack

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides extensive command-line payloads for SDR utilities like hackrf_transfer, rtl_power, and gr-gsm across multiple files, enabling hardware interaction and signal capture.
  • [DYNAMIC_EXECUTION]: Technical guides such as the GPS and Sub-GHz guides contain embedded Python scripts for signal processing and decoding. These scripts are intended for runtime execution by the agent using libraries like numpy and scipy.
  • [PRIVILEGE_ESCALATION]: The skill methodology includes the use of sudo for installing system-level dependencies and configuring hardware interfaces into monitor mode, which are required operations for radio reconnaissance.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from radio transmissions via tools like gr-gsm and dump1090. As documented in the GSM and Satellite guides, the system lacks explicit boundary markers or sanitization for these external inputs, which could potentially be used to influence agent behavior through adversarial radio signals.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:36 PM