better-notion
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill manages authentication by reading a Notion API key from a local configuration file at ~/.config/notion/api_key. This is a standard and acceptable practice for local tool configuration, and the key is used exclusively for requests to the well-known api.notion.com domain.
- [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to read content from Notion pages and databases, which creates a surface for indirect prompt injection. 1. Ingestion points: Notion API endpoints for retrieving pages and block children as documented in SKILL.md. 2. Boundary markers: None mentioned in the instructions to distinguish external data from system prompts. 3. Capability inventory: Commands for searching, reading, updating, and deleting Notion content (SKILL.md). 4. Sanitization: The instructions do not specify any validation or sanitization for the data retrieved from Notion.
Audit Metadata