claude-video
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external, untrusted video transcripts and keyframes.
- Ingestion points: External video URLs (YouTube, Instagram, TikTok) and local file paths referenced in SKILL.md.
- Boundary markers: The skill instructions do not define specific delimiters or instructions to prevent the agent from following commands that might be embedded in the video transcripts.
- Capability inventory: The workflow involves fetching transcripts and frames via the
/cloud-videocommand mentioned in SKILL.md. - Sanitization: No sanitization or validation of the ingested transcript content is described in the skill documentation.
Audit Metadata