speed-up-repetitive-parts
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute setup commands (e.g.,
claude mcp add) to register the Clueso connector on the host system.\n- [EXTERNAL_DOWNLOADS]: The skill connects to the Clueso domain (clueso.io) to configure the required MCP tool for editing workflows.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data via video projects and raw recordings.\n - Ingestion points: User-provided project links or screen recording uploads (SKILL.md).\n
- Boundary markers: The instructions lack specific delimiters or instructions to ignore embedded commands within processed media content.\n
- Capability inventory: The skill uses tools to modify video structure, regenerate narration, and export project files.\n
- Sanitization: No sanitization or content validation for ingested media or narration scripts is mentioned.
Audit Metadata