qemu-tcg-frontend
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The instructions specify the use of shell commands such as
git,meson, andninjato manage the build process and run tests. This behavior is standard and necessary for the skill's purpose of modifying and verifying QEMU source code. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data including guest ISA specifications and QEMU source files which could potentially contain adversarial content designed to influence the agent's behavior.
- Ingestion points: The agent reads from
target/directories,MAINTAINERSfiles, and external guest binaries used for functional testing. - Boundary markers: The skill does not implement specific delimiters to isolate untrusted source code or ISA documentation from the agent's primary instructions.
- Capability inventory: The agent has the capability to execute build scripts and test harnesses in the local shell environment.
- Sanitization: The skill includes a mitigation strategy by requiring the pinning of SHA-256 hashes for any external assets used in functional tests to ensure their integrity.
Audit Metadata