skills/light0305/light/light-citation/Gen Agent Trust Hub

light-citation

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to well-known academic and metadata services, including Crossref, OpenAlex, Semantic Scholar, and Unpaywall. These requests are used to verify DOI authenticity, check for retracted papers, and retrieve citation metadata.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute provided Python scripts (verify_refs.py, doi_to_any.py, and verify_citation_edge.py) for automated citation processing and validation. These scripts are self-contained and utilize only the Python standard library to interact with external APIs.
  • [DATA_EXFILTRATION]: No sensitive data access or exfiltration patterns were detected. Network operations are strictly limited to the public academic repositories mentioned above, and the author's email is used as per API guidelines for the Crossref 'polite pool' to identify the source of requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 05:04 AM
Security Audit — agent-trust-hub — light-citation