notion

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the curl utility to perform HTTP requests to the Notion API.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with api.notion.com, which is the official endpoint for the well-known Notion service.
  • [DATA_EXFILTRATION]: The skill retrieves a Notion API key from ~/.config/notion/api_key and transmits it to api.notion.com via an Authorization header for legitimate authentication.
  • [PROMPT_INJECTION]: The skill processes data from Notion pages and blocks, creating a surface for indirect prompt injection attacks.
  • Ingestion points: Content is retrieved from external sources via GET /v1/pages/{page_id} and GET /v1/blocks/{page_id}/children as documented in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are provided to the agent to help it distinguish between data and instructions within the ingested Notion content.
  • Capability inventory: The skill allows the agent to perform extensive operations including searching, reading, creating, and modifying Notion content using curl.
  • Sanitization: The instructions do not describe any mechanisms for validating or sanitizing the content fetched from Notion before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 08:53 AM
Security Audit — agent-trust-hub — notion