hardware-security
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes processing external data from hardware debug interfaces (UART console output) which may contain instructions that could influence agent behavior if not properly delimited.\n
- Ingestion points: UART serial console output (SKILL.md).\n
- Boundary markers: None present.\n
- Capability inventory: Command execution via screen, openocd, and flashrom (SKILL.md).\n
- Sanitization: None present.
Audit Metadata