video-wan
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation and JSON reference data. It contains no executable scripts, binaries, or automated tasks that could compromise a system.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection. 1. Ingestion points: Processes documents and web content as creative context. 2. Boundary markers: The guide in references/guide.md instructs treating embedded instructions as untrusted source content and ignoring permissions to execute commands or browse unrelated sites. 3. Capability inventory: No script-based capabilities, file-writing tools, or direct execution primitives are provided within the skill files. 4. Sanitization: Instructs the agent to extract intended stories, identify claims, and remove private material before further processing.
- [EXTERNAL_DOWNLOADS]: The skill references official Alibaba Cloud documentation and service surfaces. These links target a well-known cloud service provider for informational purposes and do not represent a security risk.
Audit Metadata