status
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes Python scripts located in the skill's subdirectory (session_manager.py, dag_analyzer.py, board_manager.py) to retrieve session information and analysis results. \n- [INDIRECT_PROMPT_INJECTION]: The skill accepts user input for the {session-id} variable and interpolates it directly into shell commands. This creates a potential attack surface for command injection if the input is not sanitized by the platform or agent. \n
- Ingestion points: {session-id} provided as a command argument in SKILL.md. \n
- Boundary markers: None (no delimiters or instructions to ignore embedded commands are present). \n
- Capability inventory: Execution of Python scripts via subprocess calls. \n
- Sanitization: No sanitization or validation of the {session-id} input is specified in the skill instructions.
Audit Metadata