script-generator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data such as TASK, TARGETS, and CREDENTIALS to generate executable scripts without specifying sanitization or boundary markers. Ingestion points: TASK, TARGETS, CREDENTIALS, and CONTEXT fields in the Request Format section of SKILL.md. Boundary markers: None present to delimit user input from script generation logic. Capability inventory: The skill generates standalone Python, PowerShell, and Bash scripts and provides a corresponding shell command (e.g., python3 OUTPUT_DIR/artifacts/task_name.py) to execute them. Sanitization: Absent; the instructions do not require validation or escaping of user-provided content before script interpolation.
- [DYNAMIC_EXECUTION]: The skill generates script files and provides the shell command lines necessary to execute them on the system. This pattern involves simple script generation from user-supplied logic.
Audit Metadata