research

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill uses requests, beautifulsoup4, and feedparser to download and parse content from external web sources and APIs. While typical for a research tool, this creates a network footprint.
  • [PROMPT_INJECTION] (MEDIUM): The skill is vulnerable to Indirect Prompt Injection (Category 8) due to the ingestion of untrusted external data.
  • Ingestion points: Functions like fetch_url, extract_page_content, fetch_feed, and search_nvd (documented in references/REFERENCE.md) ingest content directly from the internet.
  • Boundary markers: Absent. There are no specified delimiters or system instructions designed to prevent the agent from obeying commands found within the retrieved content.
  • Capability inventory: The skill possesses network access (requests) and produces research findings that directly influence the agent's context and decision-making via generate_report.
  • Sanitization: Absent. The documentation does not indicate any filtering, sanitization, or safety-checking of the content retrieved from external URLs or RSS feeds.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 12:50 PM