video-to-article
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted video transcription data which could contain malicious instructions.\n
- Ingestion points: Source video transcription analysis (Step 2).\n
- Boundary markers: No specific boundary markers or 'ignore' instructions are defined for the transcript input.\n
- Capability inventory: Writing to Clueso projects, capturing video frames, and image generation.\n
- Sanitization: No content sanitization is described.\n- [COMMAND_EXECUTION]: The skill instructions include suggesting the execution of shell commands for tool setup.\n
- Evidence: Instructions for running
claude mcp addfor Claude Code or plugin installation commands for ChatGPT and Codex.\n- [EXTERNAL_DOWNLOADS]: The skill requires connecting to an external service for its core functionality.\n - Evidence: Reference to
https://connect.clueso.io/mcpas the transport address for the Model Context Protocol server.
Audit Metadata