prompt-optimizer
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to run local Python scripts (
scripts/evaluate.py,scripts/optimize.py) using shell commands that incorporate user-supplied prompt text as arguments. This pattern could be vulnerable to command injection if the strings are executed in a shell without proper sanitization. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, which may contain instructions designed to subvert agent behavior. Ingestion points: User-provided prompts intended for analysis and enhancement (SKILL.md). Boundary markers: No explicit markers are used to isolate the untrusted input. Capability inventory: Execution of local analysis scripts and generation of transformed prompt content (SKILL.md). Sanitization: The instructions do not specify any validation or escaping of the user-provided prompt before processing.
Audit Metadata