check-citations

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs its stated function of academic citation verification without any detected security risks or malicious behaviors.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with reputable academic services (CrossRef, Semantic Scholar, and OpenAlex) via their public APIs. These are well-known services and the network activity is strictly scoped to querying citation metadata.
  • [DATA_EXFILTRATION]: Metadata from user-specified bibliography files (titles and DOIs) is transmitted to scholarly databases for verification. This is a documented functional requirement and does not involve sensitive personal information, credentials, or system files.
  • [COMMAND_EXECUTION]: The provided Python scripts parse local files and perform network requests using the standard requests library. No unsafe dynamic execution (such as eval() or exec()) or unauthorized shell commands were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 11:31 AM
Security Audit — agent-trust-hub — check-citations