captions-and-clipping
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define workflows for processing external video content and transcripts, which are untrusted data sources. * Ingestion points: User-provided long-form videos, webinars, and podcasts as referenced in
SKILL.mdandreferences/clip-and-caption-recipes.md. * Boundary markers: The skill lacks explicit delimiters for separating external data from agent instructions, although it explicitly states that external content like comments or DMs should be treated as content and not commands. * Capability inventory: This skill serves as a documentation and routing guide; it does not contain direct file-write, network, or subprocess execution scripts. * Sanitization: No specific sanitization or filtering of external transcript data is mentioned. - [NO_CODE]: The skill distribution is composed entirely of Markdown documentation, JSON metadata, and reference guides. It does not include Python, JavaScript, Shell scripts, or other executable binaries.
Audit Metadata