notion-pages

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Communicates with the official Notion API (api.notion.com). This interaction is the primary purpose of the skill and targets a well-known service.\n- [DATA_EXFILTRATION]: Accesses the configuration file ~/.config/notion/config.json to retrieve the Notion API token. This is a legitimate use of locally stored credentials required for authentication with the Notion service.\n- [PROMPT_INJECTION]: The skill reads external content from Notion pages, which constitutes a surface for indirect prompt injection if a page contains malicious instructions.\n
  • Ingestion points: Content and titles are ingested from Notion via the cmd_get and cmd_search functions in scripts/notion_pages.py.\n
  • Boundary markers: None are implemented; data is returned as plain markdown-formatted text to the agent.\n
  • Capability inventory: The skill can search, read, create, update, and append Notion pages. It does not have access to the local shell or file system writes.\n
  • Sanitization: Content is converted from Notion's block format to markdown text without any filtering for instructions or malicious patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 08:42 PM