Skills
skills/apsp-ag/agent-skills/wsl-embedded-debugging/Gen Agent Trust Hub

wsl-embedded-debugging

Warn

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Instructions in SKILL.md and references/wsl-windows-embedded-troubleshooting.md advise the agent to bypass security restrictions, such as switching to a 'less restricted mode' or using 'full-access sandboxing' if execution is blocked by policy.\n- [COMMAND_EXECUTION]: The script scripts/run_windows_embedded.sh is an execution wrapper that runs arbitrary commands passed as arguments.\n- [PROMPT_INJECTION]: The skill captures untrusted firmware logs, which constitutes an indirect prompt injection surface. Ingestion points: scripts/run_windows_embedded.sh redirects output to log files. Boundary markers: None. Capability inventory: Subprocess execution of Windows host binaries. Sanitization: None.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 17, 2026, 09:50 AM