video-analyzer
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to access and analyze video content from external platforms based on user-provided URLs. This is the primary function of the skill and targets well-known video-hosting services.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes metadata and content from external videos. However, the instructions use highly structured reporting templates which serve as constraints on the agent's behavior, mitigating the risk of the agent following instructions embedded within video content.
- Ingestion points: External video URLs and their associated metadata (title, description, tags) processed via user requests in
SKILL.md. - Boundary markers: None explicitly defined to encapsulate external data.
- Capability inventory: File system write access used to save analysis reports to
./output/reports/. - Sanitization: Not explicitly defined in the instructions.
Audit Metadata