console-levels
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from PlayStation 2 disc files, archives, and runtime logs to map and unpack level data. While the script implements sanitization for extracted filenames and verifies that file operations remain within the designated output directory, the agent's behavior could be influenced by malicious content embedded within these processed files.\n
- Ingestion points: Binary data from game archives and text from runtime logs are read by datascan.py (e.g., in cmd_unpack and cmd_files).\n
- Boundary markers: The output of the file parsing is presented to the agent without specific delimiters or warnings to ignore embedded instructions.\n
- Capability inventory: The skill can perform file-write operations through datascan.py unpack and repack commands.\n
- Sanitization: The script includes _safe_name to sanitize filenames and _inside to prevent directory traversal during the repacking process.\n- [COMMAND_EXECUTION]: The skill executes a local Python script, datascan.py, to perform file analysis and manipulation. This script is limited to the Python standard library and does not invoke external shell commands or network services.
Audit Metadata