brand-asset
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of a local script,
docs/creative_cli.py, which is used to communicate with the Fal.ai image generation service. This execution is triggered when a user-supplied API key is detected.\n- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it incorporates user-supplied brand information and style anchors into prompts sent to an image generation model.\n - Ingestion points: Intake process in
SKILL.mdwhere brand name, asset types, and style anchors are collected.\n - Boundary markers: None identified to separate user-provided data from the prompt templates.\n
- Capability inventory: Subprocess execution of
docs/creative_cli.pyas defined in the Workflow section ofSKILL.md.\n - Sanitization: No input validation or sanitization for brand names or creative descriptions is documented in the instructions.
Audit Metadata