zephyr-debug
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes system utilities such as GDB, OpenOCD, J-Link, and QEMU to perform debugging tasks.
- [COMMAND_EXECUTION]: The instructions direct the agent to generate and execute Bash, Python, and GDB scripts stored in a task-specific workspace directory (".zephyr-skills/").
- [PROMPT_INJECTION]: The skill processes untrusted external data such as logs and third-party libraries; it manages this surface by isolating artifacts in a dedicated workspace folder and excluding them from version control, though no specific sanitization for processed content is described.
Audit Metadata