nby-notion-smart-categorize
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes curl to interact with the Notion REST API and python3 to parse JSON responses into readable text. These operations are confined to official Notion API endpoints.
- [EXTERNAL_DOWNLOADS]: Fetches external web content via r.jina.ai for link analysis when a Notion page contains primarily bookmarks. Jina AI is a recognized service used for extracting content in an LLM-friendly format.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it analyzes untrusted data from Notion page blocks and external websites. 1. Ingestion points: Notion page block content (SKILL.md) and web content fetched through Jina AI (SKILL.md). 2. Boundary markers: No explicit delimiters or instructions for the agent to ignore embedded commands within the fetched content are present. 3. Capability inventory: The skill possesses the ability to move Notion pages via API, execute shell commands, and perform network fetches. 4. Sanitization: No content validation or sanitization mechanisms are described.
- [SAFE]: The skill implements secure credential management by guiding the user to store their Notion API token in local .env files rather than hardcoding secrets or passing them in plaintext.
Audit Metadata