recordings-to-training-module

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill includes instructions for connecting the 'clueso-mcp' tool using the endpoint https://connect.clueso.io/mcp. This is a vendor-owned resource belonging to 'clueso-ai' and represents standard configuration for the skill's intended functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to transcribe and analyze user-provided video recordings to ensure consistency and alignment with learning goals. This creates a potential surface for indirect prompt injection if a video transcript contains malicious instructions designed to influence the agent's behavior during analysis.
  • Ingestion points: Video recordings uploaded by the user and the resulting audio transcriptions processed in the 'Workflow' section (Step 4).
  • Boundary markers: The instructions do not define specific delimiters or guardrails to differentiate between transcript content and agent instructions.
  • Capability inventory: The skill utilizes the clueso-mcp tool for video project management, clip arrangement, and automated narration generation.
  • Sanitization: No explicit sanitization or filtering of transcript content is performed; the agent is directed to analyze the text for pacing and clarity.
  • [EXTERNAL_DOWNLOADS]: The skill provides the user with commands and links to install the Clueso MCP connector from connect.clueso.io and view setup documentation at help.clueso.io. These references target official vendor infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 06:28 AM