soft-skills-scenario-video

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to connect a Model Context Protocol (MCP) server at https://connect.clueso.io/mcp. This is a vendor-owned resource from Clueso, used for the skill's primary video generation functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided data such as policy documents and incident reports to generate video scripts, creating a surface for indirect prompt injection.
  • Ingestion points: Processes external policy sources and incident descriptions provided by the user in the 'Inputs' section of SKILL.md.
  • Boundary markers: The skill does not define specific XML or markdown delimiters for these inputs, but relies on a human-in-the-loop workflow.
  • Capability inventory: The skill generates scripts and interfaces with the Clueso MCP for video scene creation and export.
  • Sanitization: The workflow incorporates strong mitigations, requiring the agent to share drafts and obtain explicit user sign-off on policy-sensitive wording and modeled responses before building or exporting the final video.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 06:28 AM