The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of processing external data from Notion.
Ingestion points: The skill retrieves untrusted data from the user's workspace using Notion:notion-search, Notion:notion-fetch, and Notion:notion-query-data-sources as defined in SKILL.md and demonstrated in examples/project-decision.md.
Boundary markers: Absent; there are no instructions or templates in the skill files that define delimiters or provide warnings to the agent to ignore instructions embedded within the fetched Notion content.
Capability inventory: The skill has significant write capabilities, including Notion:notion-create-pages, Notion:notion-update-page, and Notion:notion-create-comment (referenced in SKILL.md and examples/project-decision.md).
Sanitization: Absent; the skill does not implement any escaping, validation, or filtering of the content retrieved from Notion before it is used to generate meeting materials.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to configure a connection to an external Notion Model Context Protocol (MCP) server.
The SKILL.md file directs the user to add the Notion MCP via the URL https://mcp.notion.com/mcp. This is a standard configuration for integrating with the official Notion service and does not represent a malicious download.

notion-meeting-intelligence