ticket-resolution-to-how-to-video
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill integrates with Clueso's Model Context Protocol (MCP) to provide video editing capabilities. It provides instructions for users to connect the official vendor endpoint at connect.clueso.io/mcp.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it analyzes untrusted video transcripts and screen content to generate new narration. While this is the primary purpose of the skill, the following indicators were identified: 1. Ingestion points: Video transcripts and screen content are analyzed in Workflow Step 2. 2. Boundary markers: The instructions do not specify delimiters or warnings to ignore instructions within the source material. 3. Capability inventory: The skill uses tools to generate narration, modify video timelines, and export projects. 4. Sanitization: No specific filtering or validation of transcribed text is mentioned before it is processed for narration rewriting.
Audit Metadata