re-automotive

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill instructs the agent to ingest and parse CAN bus log files, diagnostic dumps, and firmware update packages, which are externally-sourced data payloads.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill instructs the agent to run commands with sudo privileges to configure virtual CAN networks, which compromises the state of the machine by modifying network interfaces and kernel modules.