skills/clueso-ai/skills/revoice-video/Gen Agent Trust Hub

revoice-video

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill points to official author resources at clueso.io for its Model Context Protocol (MCP) server and documentation. These are considered safe vendor resources.\n- [COMMAND_EXECUTION]: Includes instructions for the agent or user to run environment configuration commands like claude mcp add. This is standard practice for setting up required tool integrations.\n- [PROMPT_INJECTION]: The skill ingests video transcript data from Clueso projects or uploaded recordings (Ingestion Point: SKILL.md). Evidence Chain: 1. Ingestion: Video transcript data; 2. Boundaries: Absent; 3. Capabilities: Script editing and visual re-alignment via Clueso MCP; 4. Sanitization: Absent (rewriting content for tone naturally mitigates instruction propagation).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 09:42 AM
Security Audit — agent-trust-hub — revoice-video