notion-api

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • Data Ingestion Surface: The skill facilitates reading content from external workspace pages and databases. It correctly identifies this as a potential surface for indirect prompt injection and includes a security note advising that retrieved data be treated as untrusted.\n
  • Ingestion points: Content is retrieved via bundled scripts for reading pages and searching the workspace.\n
  • Boundary markers: A clear security warning is present in the documentation to prevent following instructions embedded in retrieved content.\n
  • Capability inventory: The skill is scoped to API interactions and local script execution.\n
  • Sanitization: The instructions emphasize treating data as inert evidence.\n- Service Integration: All network activity is directed to the official service domain (api.notion.com), and authentication is handled through environment variables managed by the runtime.\n- Bundled Scripts: The provided scripts for searching and reading data are standard implementations using common utilities, with appropriate error handling for service communication.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 11:18 PM
Security Audit — agent-trust-hub — notion-api