dayz-3d-viewer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external binary asset files, creating a potential surface for indirect injection attacks. 1. Ingestion points: The skill ingests .p3d models, .paa textures, and .rvmat material files provided by the user. 2. Boundary markers: No explicit prompt delimiters or instructions are provided to the agent to treat asset data as untrusted. 3. Capability inventory: The skill performs file system operations, generates HTML, and executes installation subprocesses. 4. Sanitization: The skill relies on third-party libraries such as Pillow, NumPy, and pygltflib to parse binary formats, reducing the risk of direct exploit.
- [COMMAND_EXECUTION]: The skill utilizes shell commands for environment setup and dependency management. Evidence: The installer script scripts/install_py3d.py uses subprocess.run to execute pip commands. SKILL.md provides several pip and apt-get commands for manual environment configuration.
- [DYNAMIC_EXECUTION]: The installer script performs dynamic verification of the local environment. Evidence: The script install_py3d.py uses importlib.import_module to inspect an existing library. It uses ast.parse and ast.literal_eval to safely analyze the library's source code for a specific security flag (IS_DAYZ_FORK) before allowing modification, ensuring a safe and verifiable installation process.
Audit Metadata