literature-review

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted content from external academic APIs which could contain malicious instructions designed to influence agent behavior.
  • Ingestion points: The scripts/lit_search.py script fetches paper titles and abstracts from Semantic Scholar, OpenAlex, Crossref, and PubMed.
  • Boundary markers: The script outputs data as a JSON object but does not include specific boundary markers or 'ignore' instructions to separate data from commands.
  • Capability inventory: The script is limited to performing network GET requests and printing to stdout; it does not have the ability to write to the file system or execute arbitrary commands.
  • Sanitization: No sanitization or filtering is applied to the abstracts or metadata retrieved from the external services.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 03:01 PM