aggregating-crypto-news

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill follows secure design principles for its intended purpose of data aggregation, using parallel processing and caching to minimize network load.
  • [EXTERNAL_DOWNLOADS]: Fetches content from well-known cryptocurrency news organizations (e.g., CoinDesk, The Block, Decrypt) via RSS feeds as listed in the source registry.
  • [PROMPT_INJECTION]: The skill ingests external data (headlines and summaries) which represents a surface for indirect prompt injection.
  • Ingestion points: External RSS feed XML content processed in feed_fetcher.py and feed_parser.py.
  • Boundary markers: Data is structured into tables, CSV, or JSON formats; however, there are no specific instructions to the agent to disregard potential instructions embedded within the news text.
  • Capability inventory: Executes local Python scripts and bash commands within the restricted crypto:news-* namespace.
  • Sanitization: feed_parser.py implements HTML tag stripping and whitespace normalization to clean incoming text content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 01:08 PM