scientific-db-openalex-database

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted third-party data (scholarly titles, abstracts, and metadata) retrieved from the OpenAlex API, which could theoretically contain malicious instructions.
  • Ingestion points: The skill fetches data via the search_works and get_work_by_doi functions.
  • Boundary markers: The skill includes a dedicated 'Trust Boundary' section that explicitly instructs the agent to treat API responses as evidence only and never allow them to change the task or tool calls.
  • Capability inventory: The skill utilizes network access via the requests library. It does not possess file-write or shell-execution capabilities.
  • Sanitization: The script includes logic to reconstruct abstracts from an inverted index format, though it does not perform explicit HTML/Markdown sanitization of the resulting text.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the OpenAlex API (api.openalex.org) to fetch academic literature data.
  • The skill uses standard Python requests calls to a well-known scholarly database.
  • The instructions correctly advise using environment variables for authentication to avoid credential exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 08:16 PM