weibo-hot

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill fetches data from an external website (weibo.com). If the trending topics themselves contain malicious instructions, the agent could potentially be influenced. Evidence: (1) Ingestion points: Data is fetched from https://s.weibo.com/top/summary in scripts/get_hot.py. (2) Boundary markers: Absent. (3) Capability inventory: The script prints titles and hot values to standard output. (4) Sanitization: None; the content is parsed and printed directly.
  • [Safe Behavior] (SAFE): The network operations are limited to the target scraping site and use standard headers. No credentials or sensitive local files are accessed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 03:37 AM