crypto-news-ns3

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions utilize the curl tool to fetch real-time cryptocurrency news and market briefings from NS3 AI API endpoints (e.g., api.ns3.ai/feed/news-data). These operations are core to the skill's functionality and are limited to fetching public news data.
  • [PROMPT_INJECTION]: The skill ingests untrusted third-party news content from over 20 media outlets via a remote API, creating a surface for indirect prompt injection.
  • Ingestion points: Data is ingested via https://api.ns3.ai/feed/news-data, https://api.ns3.ai/feed/news-ranking, https://api.ns3.ai/feed/today-summary, and https://api.ns3.ai/feed/news-flash.
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are used to wrap the fetched news content within the prompt.
  • Capability inventory: The skill uses curl to perform network operations to retrieve the data.
  • Sanitization: The skill describes a server-side AI classification pipeline (Stages 1-4) that filters promotional noise and routine content before delivery.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 01:06 AM
Security Audit — agent-trust-hub — crypto-news-ns3