Skills
Navigate
skills/cinience/alicloud-skills/aliyun-wan-videoedit/Gen Agent Trust Hub

aliyun-wan-videoedit

Pass

Audited by Gen Agent Trust Hub on Oct 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text prompts and media URLs that are directly incorporated into API requests, which subsequently triggers file system writes of the task results.
  • Ingestion points: The prompt string and media object array in SKILL.md and scripts/edit_video.py.
  • Boundary markers: None observed; input data is mapped directly to the JSON payload for the API call.
  • Capability inventory: The skill performs network requests to dashscope.aliyuncs.com and executes file writes to the output/aliyun-wan-videoedit/ directory.
  • Sanitization: No validation or escaping of the user-provided prompt or media URLs is performed.
  • [EXTERNAL_DOWNLOADS]: The skill installation instructions include downloading the dashscope SDK from the official Python package registry.
  • Evidence: python -m pip install dashscope located in the Prerequisites section of SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 15, 2026, 08:17 AM