hardware-radio-sdr

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of radio frequency (RF) captures.
  • Ingestion points: Raw I/Q samples are captured from the environment via SDR hardware and saved to files (e.g., capture.cs8) for analysis in tools like URH and Inspectrum (SKILL.md, Method steps 3 and 4).
  • Boundary markers: The instructions mandate the use of scope.txt and tradecraft-scope-roe to define legal and technical boundaries, and emphasize verifying transmission authorization.
  • Capability inventory: The skill utilizes subprocess calls to external binaries including hackrf_transfer, gqrx, rtl_power, and urh (SKILL.md, Method steps 2, 3, 4, and 6).
  • Sanitization: The skill includes instructions to sanitize evidence by removing sensitive frequency and location combinations before reporting (SKILL.md, Gotchas).
  • [EXTERNAL_DOWNLOADS]: The skill references a repository by zhaoxuya520 (github.com/zhaoxuya520/reverse-skill) as a source for its content attribution under the MIT License. This is a passive documentation reference and does not involve automated downloads or execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 10:30 AM