creative-strategist
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow includes an optional step to execute a local Python script (docs/creative_cli.py) via the shell to generate test images based on the established brand direction.
- [INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection because it ingests untrusted user data and interpolates it into a shell command without sanitization.
- Ingestion points: Data gathered from users in the 'Intake' section (brand, audience, adjectives) in SKILL.md.
- Boundary markers: None identified in the shell command construction.
- Capability inventory: Execution of docs/creative_cli.py with shell expansion.
- Sanitization: No sanitization or shell-escaping is performed on the representative subject or style block variables.
- [DYNAMIC_EXECUTION]: The skill dynamically assembles a style block and incorporates it into a shell command for execution at runtime.
Audit Metadata