proposal-lit-search

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill handles API keys for academic services (OpenAlex, Crossref). It implements best practices by requesting the agent to store these in an api-keys.env file and explicitly requires adding this file to .gitignore to prevent credential exposure in shared repositories.
  • [EXTERNAL_DOWNLOADS]: The scripts perform network requests to well-known academic API providers including arXiv, Crossref, DBLP, OpenAlex, Semantic Scholar, and OpenCitations. These are standard integrations for the skill's primary purpose and use appropriate rate-limiting and User-Agent headers.
  • [COMMAND_EXECUTION]: The skill uses Python scripts to perform federated searches. While importlib.import_module() is used in search.py and snowball.py, it is restricted to loading the internal source client modules (arxiv, dblp, etc.) defined in the local scripts directory.
  • [PROMPT_INJECTION]: Instructions include clear guidance on how the agent should handle networking failures and how to judge academic relevance, with no attempts to bypass safety filters or override system instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 08:17 AM
Security Audit — agent-trust-hub — proposal-lit-search