openalex-database

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill connects to the OpenAlex API at https://api.openalex.org to retrieve scholarly literature data. This is a well-known and reputable public service for bibliometric data. It also recommends installing the well-known requests library via uv pip.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external source, creating a potential surface for indirect prompt injection.
  • Ingestion points: Scholarly titles, abstracts, and metadata are retrieved from the OpenAlex API via scripts/openalex_client.py and scripts/query_helpers.py.
  • Boundary markers: No explicit delimiters or instructions to ignore potential commands embedded in the retrieved scholarly content are implemented.
  • Capability inventory: The skill utilizes network requests (requests.get) and provides examples for local file system writes (CSV export in SKILL.md).
  • Sanitization: Content from the API is processed as raw text without specific sanitization for agent instructions before being presented to the model or written to storage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 11:43 AM
Security Audit — agent-trust-hub — openalex-database