The Agent Skills Directory

[SAFE]: The skill promotes secure authentication practices by instructing the use of environment variables (NOTION_TOKEN) instead of hardcoded credentials.
[SAFE]: The skill relies on official, well-known libraries (@notionhq/client and notion-client) for API interaction.
[SAFE]: The primary function of the skill is to provide PII redaction and data minimization, which are defensive security practices that reduce the overall attack surface.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external Notion pages, which is a potential vector for indirect prompt injection. However, the risk is minimized by the skill's focus on structured data and sanitization.
Ingestion points: Data is fetched from Notion databases in references/pii-detection.md and references/compliance-patterns.md.
Boundary markers: None explicitly implemented for prompt isolation.
Capability inventory: Uses Read, Write, and Edit tools for interacting with Notion.
Sanitization: Implements regex-based PII scanning and property redaction routines to sanitize content.

notion-data-handling