notion-research-documentation

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes content fetched from external Notion pages, creating a surface for indirect prompt injection. Malicious instructions within a processed page could attempt to hijack the agent's behavior during synthesis or report generation.\n
  • Ingestion points: The skill utilizes Notion:notion-fetch to retrieve content from arbitrary Notion pages as part of its core workflow.\n
  • Capability inventory: The skill has the capability to search, fetch, create, and update pages within the connected Notion workspace.\n
  • Boundary markers: There are no explicit instructions or delimiters defined to isolate fetched content from the agent's internal reasoning or system instructions.\n
  • Sanitization: No sanitization or validation logic is defined for the fetched Notion content before it is processed by the agent.\n- [EXTERNAL_DOWNLOADS]: The skill configuration and setup instructions reference a remote Notion MCP server.\n
  • Evidence: SKILL.md and agents/openai.yaml reference https://mcp.notion.com/mcp as the MCP server URL.\n
  • Context: The referenced URL targets the official Notion domain, which is a well-known and trusted service provider for the skill's intended functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 01:16 AM