verify-citation

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The script performs network requests to api.crossref.org and api.openalex.org. These are well-known academic metadata services used to verify citation existence and retrieve abstracts.
  • [COMMAND_EXECUTION]: The skill requires executing a Python script (verify_citation.py) which performs file system operations, including reading BibTeX files and reading/writing CSV summary tables.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources and local files that is then presented to the agent context.
  • Ingestion points: The script reads content from local BibTeX files (.bib) and retrieves article titles and abstracts from public APIs (CrossRef and OpenAlex).
  • Boundary markers: No explicit delimiters or instructions are used to separate the retrieved abstract text from the agent's instructions.
  • Capability inventory: The script has capabilities to read/write local files and perform network GET requests.
  • Sanitization: The script performs normalization for metadata comparison (lowercase, removing special characters) but passes raw abstract text (up to 1500 characters) directly to the agent's context without sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 09:21 AM
Security Audit — agent-trust-hub — verify-citation