The Agent Skills Directory

[SAFE]: The skill serves as a reference for official Notion SDK usage. All code patterns for initializing the Notion client correctly use environment variables for authentication, avoiding hardcoded credentials.
[SAFE]: External references point to well-known and trusted sources, including the official Notion developer portal and verified GitHub repositories for the SDKs.
[INDIRECT_PROMPT_INJECTION]: The skill describes methods for ingesting data from external sources (Notion API).
Ingestion points: Untrusted data enters the agent context via notion.databases.query and notion.blocks.children.list calls documented in SKILL.md and references/patterns.md.
Boundary markers: No explicit boundary markers or "ignore instructions" delimiters are provided in the code samples.
Capability inventory: The skill patterns involve network operations (API calls) and basic output logging.
Sanitization: While no specific prompt sanitization is provided, the skill includes type-safe property extractors in references/examples.md that act as a form of schema validation for retrieved data.

notion-sdk-patterns