add-ai-captions

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to connect to a Model Context Protocol (MCP) server at https://connect.clueso.io/mcp. This is a vendor-provided endpoint required for the skill to interact with the Clueso platform.
  • [COMMAND_EXECUTION]: Provides standard configuration commands for various environments (e.g., claude mcp add --transport http Clueso https://connect.clueso.io/mcp) to enable the necessary platform integrations.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-uploaded media to generate transcripts, which serves as a potential surface for indirect prompt injection if the audio contains malicious instructions.
  • Ingestion points: User-supplied videos and audio files are processed for transcription in the 'Transcribe the existing audio' step.
  • Boundary markers: The instructions do not define specific delimiters to isolate transcribed text from system instructions.
  • Capability inventory: The skill utilizes tools to read project metadata, create and style text elements, render preview frames, and execute video exports.
  • Sanitization: No explicit filtering or validation of the generated transcript content is mentioned prior to creating caption elements.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 09:42 AM
Security Audit — agent-trust-hub — add-ai-captions