literature-search

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The implementation is restricted to Python's standard library, effectively eliminating supply-chain risks from unverified third-party packages.
  • [SAFE]: API keys are managed through a local keys.env file with restrictive file permissions (0600). The tool is designed to only report the presence of keys as booleans, preventing the actual secret values from appearing in the agent's conversation history.
  • [SAFE]: Network activity is confined to well-known academic APIs (arXiv, Semantic Scholar, OpenAlex, OpenRouter, and bgpt.pro). These operations are used solely for retrieving scholarly metadata and full-text papers.
  • [SAFE]: File system interactions, such as caching and downloading PDFs, use sanitized identifiers to prevent path traversal or other unauthorized file access.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 04:42 PM
Security Audit — agent-trust-hub — literature-search