base-academic-search

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill provides standard documentation and code for interacting with a well-known academic search engine (BASE). All network requests are directed to official BASE API endpoints at api.base-search.net.
  • [PROMPT_INJECTION]: Analyzed for indirect prompt injection surface related to processing external search results.
  • Ingestion points: Metadata and abstracts are fetched from the BASE API in the Python helper script in SKILL.md.
  • Boundary markers: None present in the provided examples.
  • Capability inventory: Network read operations using the requests library in SKILL.md.
  • Sanitization: Not demonstrated in the example code; however, the data processed is academic metadata which is a low-risk source.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 10:16 PM