web-content-fetcher

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a wrapper for the platform's internal WebIntelligenceRouter, facilitating the transition to common web-reading pipelines while maintaining backward compatibility.
  • [SAFE]: Input validation is properly implemented in scripts/fetch.py. The _parse_max_chars function ensures that character limits are handled as integers with safe defaults, preventing potential type-related errors.
  • [SAFE]: No dangerous sinks or execution patterns were detected. The skill does not use eval(), exec(), or subprocess calls. It relies on high-level internal abstractions for network operations.
  • [SAFE]: The skill handles external URLs as its primary function, which is a standard use case for web content retrieval tools. It does not exhibit patterns of data exfiltration or unauthorized file access.
  • [SAFE]: Dependencies listed in the manifest (requests, html2text) are industry-standard libraries for HTTP requests and HTML-to-Markdown conversion, matching the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 06:34 AM