qemu-linux-boot

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill constructs and executes QEMU system commands and Git utility commands for workspace and boot process management. Evidence includes instructions to construct copy-pasteable commands and use git rev-parse to modify repository-local exclusion files in SKILL.md.\n- [PROMPT_INJECTION]: The skill processes external data from QEMU console logs and firmware output, which represents an indirect prompt injection surface. An attacker providing a malicious kernel or firmware image could embed instructions in the console output to influence the agent's behavior during log analysis.\n
  • Ingestion points: Full console logs and serial markers from various boot stages in SKILL.md.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided for the log analysis phase.\n
  • Capability inventory: The agent can execute system commands and modify local Git configurations.\n
  • Sanitization: The instructions do not include steps to sanitize or filter the captured console logs before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 01:32 AM
Security Audit — agent-trust-hub — qemu-linux-boot