news-aggregator-skill

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes a Python script (fetch_news.py) to perform news aggregation and content fetching.
  • [EXTERNAL_DOWNLOADS]: Fetches headlines and full article content from well-known domains such as ycombinator.com, github.com, producthunt.com, and 36kr.com to provide real-time updates.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it processes untrusted text from external websites.
  • Ingestion points: Fetches titles and full article content from external domains in scripts/fetch_news.py.
  • Boundary markers: SKILL.md does not include specific delimiters or instructions to ignore potential commands embedded in news articles.
  • Capability inventory: The skill uses python3 and requires network connectivity to function.
  • Sanitization: The script performs basic HTML cleanup but lacks logic to detect or filter adversarial prompt instructions in the processed text.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:44 PM
Security Audit — agent-trust-hub — news-aggregator-skill