multilingual-captions
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes potentially untrusted content from video transcripts and user-provided glossary terms.
- Ingestion points: Transcript data derived from audio analysis and user glossary inputs in SKILL.md.
- Boundary markers: None present to distinguish transcript content from system instructions.
- Capability inventory: Access to clueso-mcp tools for project duplication, element modification, and video export.
- Sanitization: No explicit sanitization or validation of the ingested text is performed before translation and placement.
- [EXTERNAL_DOWNLOADS]: The skill documentation provides setup instructions involving external URLs (connect.clueso.io, help.clueso.io) and specific plugin identifiers for platform integration. These resources are associated with the skill's primary vendor.
Audit Metadata