routeros-ethernet
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from network device outputs which could potentially include adversarial instructions.
- Ingestion points: Data is read from device commands such as /interface/ethernet/monitor and /interface/pwr-line/monitor (specified in SKILL.md, references/guides/wired-connections/ethernet.md, and references/guides/wired-connections/pwr-line.md).
- Boundary markers: The instructions do not define delimiters or instruct the agent to ignore instructions within device-returned strings.
- Capability inventory: The skill includes commands to modify device configurations, such as /interface/ethernet/set (specified in SKILL.md) and /interface/pwr-line/configure (specified in references/guides/wired-connections/pwr-line.md).
- Sanitization: There is no evidence of output sanitization or validation of data retrieved from the router.
Audit Metadata