skills/wbh604/uzi-skill/trap-detector/Gen Agent Trust Hub

trap-detector

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple external sources, primarily through web searches for stock-related keywords across social media platforms like WeChat, Douyin, and Bilibili.
  • Ingestion points: Untrusted content is retrieved for 8 risk signals defined in SKILL.md and references/eight-signals.md, including 'low-quality accounts' and 'templated recommendation scripts'.
  • Boundary markers: The instructions do not define any delimiters or provide 'ignore embedded instructions' warnings for the agent when handling the retrieved content.
  • Capability inventory: The skill has access to financial analysis tools including fetch_financials, fetch_sentiment, and fetch_kline to verify stock data.
  • Sanitization: No sanitization or validation mechanisms are implemented to filter malicious content from the ingested search results before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:48 AM
Security Audit — agent-trust-hub — trap-detector