soybean-headless

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and API reference metadata from the official vendor domain ui.soybeanjs.cn.
  • [INDIRECT_PROMPT_INJECTION]: The skill design includes an indirect prompt injection attack surface. 1. Ingestion points: The agent is instructed to load content from https://ui.soybeanjs.cn/llms.txt and component documentation files. 2. Boundary markers: The skill does not define boundary markers or specific safety instructions for the agent to follow when processing this external data. 3. Capability inventory: The agent is tasked with using the retrieved information to generate Vue code and suggest project commands. 4. Sanitization: The instructions do not include any validation or sanitization steps for the data fetched from the remote documentation routes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 05:06 AM