Skills
Navigate
skills/qwenlm/qwen-mm-plugins/qwen-mm-plugins-search/Gen Agent Trust Hub

qwen-mm-plugins-search

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external websites via web search and extraction tools, which could contain malicious prompts.
  • Ingestion points: Content from web_extractor and web_search tools in SKILL.md and references/video_search.md.
  • Boundary markers: No specific markers are used to delimit external content from system instructions.
  • Capability inventory: The plugin suite has access to video reading (read_video), frame saving (save_view), and vision chat capabilities.
  • Sanitization: No specific filtering or sanitization of external content is documented.
  • [DATA_EXFILTRATION]: The image_search tool is designed to upload local media to a third-party hosting service for reverse image identification.
  • Evidence: Documentation in references/video_search.md confirms that local images are uploaded to the third-party host uguu.se.
  • Mitigation: The skill mandates obtaining user consent and passing a specific flag (allow_public_upload=true) before any upload occurs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:32 AM