mage-youtube-thumbnail
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes the official Mage MCP connector and specific image generation models (Mango 4 and GPT Image 2.5 Flare) provided by the author. All external references are to the vendor's own domain (mcp.mage.space), and no malicious command execution or unauthorized data access patterns were found.
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided input, such as video titles and headline text, which are used to generate visual content. This constitutes a standard ingestion surface for generative tasks.
- Ingestion points: User-supplied video topics, titles, and headline text in SKILL.md.
- Boundary markers: The skill uses structured prompt construction (e.g., CHARACTER N labels and specific quoted text instructions) to help delineate user content from system instructions.
- Capability inventory: The skill uses generate and gpt_image_2 tools to create and modify image assets.
- Sanitization: No manual sanitization logic is present; the skill relies on the safety protocols of the underlying image generation models.
Audit Metadata