tella-title-cards
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest external data from video metadata and chapters via the
get_videotool to generate text for opening, chapter, and end cards. This represents a potential surface for indirect prompt injection if the source video content contains malicious instructions. - Ingestion points: Video metadata and chapter information retrieved via the
get_videotool (referenced inSKILL.md). - Boundary markers: No specific delimiters or safety instructions are provided to distinguish between the extracted text and the agent's instructions.
- Capability inventory: The skill utilizes Tella MCP tools to modify video timelines and design graphics.
- Sanitization: The instructions do not specify any sanitization or validation of the text content extracted from the video before it is used to generate title cards.
Audit Metadata