aesthetic
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill ingests untrusted data from external inspiration websites (Dribbble, Behance, etc.) via chrome-devtools and processes it using ai-multimodal. This creates a surface where a malicious website could attempt to influence the agent's behavior. 1. Ingestion points: SKILL.md Workflow 1 involves browsing and capturing screenshots of external websites. 2. Boundary markers: Absent; there are no specific instructions to ignore embedded commands in the analyzed content. 3. Capability inventory: Uses chrome-devtools for web interaction, ai-multimodal for image analysis, and media-processing (FFmpeg/ImageMagick) for media manipulation. 4. Sanitization: No sanitization or validation of the fetched web content or screenshots is mentioned.
- [Command Execution] (LOW): The skill references the use of command-line tools such as gallery-dl, FFmpeg, and ImageMagick for media processing tasks. These represent a standard execution surface for processing potentially untrusted external media files.
Audit Metadata