ig-carousel
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines clear, benign instructions for an AI agent to assist in content creation for Instagram. The technical suggestions for rendering HTML to images are provided as workflow examples rather than executable commands.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied ideas and lists to format them into slides. While this represents a standard attack surface for indirect injection, the risk is inherent to the skill's primary function of content transformation. 1. Ingestion points: User-provided text, ideas, or lists mentioned in the skill description. 2. Boundary markers: None present in the instructions. 3. Capability inventory: The skill describes building HTML files and printing them to PDF/images using external tools. 4. Sanitization: No explicit sanitization or validation of user input is defined.
Audit Metadata