scientific-literature-search

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches content from recognized academic sources and publishers such as nature.com, arxiv.org, and api.crossref.org. These references are appropriate for the primary purpose of literature searching.
  • [COMMAND_EXECUTION]: The workflow involves using common Python packages for data retrieval and processing, including biopython, arxiv, and pypdf. These are legitimate tools in a research context.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted content from the web and PDF files using trafilatura and pypdf (Workflow Step 5 in SKILL.md). While the skill lacks explicit boundary markers or sanitization, this risk is inherent to literature analysis tasks and is mitigated by the intended use-case of research synthesis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:35 PM
Security Audit — agent-trust-hub — scientific-literature-search