The Agent Skills Directory

[COMMAND_EXECUTION]: Uses curl and jq to interact with the Notion API. It also writes temporary data to /tmp/notion_request.json to handle JSON payloads for POST and PATCH requests.- [DATA_EXFILTRATION]: Accesses Notion workspace data by making network requests to the well-known service api.notion.com. Authentication is handled via the $NOTION_TOKEN environment variable, following standard security practices for credential management.- [PROMPT_INJECTION]: Contains an indirect prompt injection surface as it retrieves content from untrusted external Notion pages and databases.
Ingestion points: The skill fetches page content, block children, and database entries using API calls in SKILL.md (e.g., Read Page with Content).
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat fetched content as data rather than instructions.
Capability inventory: The skill has access to network operations via curl and local file writes to /tmp.
Sanitization: There is no evidence of sanitization or filtering applied to the data retrieved from Notion before it enters the agent context.

notion