The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: the skill fetches and processes content from a user's Notion workspace using notion-search and notion-fetch to generate meeting materials. This creates a surface for indirect prompt injection if the ingested documents contain adversarial instructions intended to influence the agent's behavior during document generation.
Ingestion points: data enters the agent context through Notion:notion-search and Notion:notion-fetch referenced in SKILL.md (Workflow Step 1) and various example files.
Boundary markers: the instructions do not specify the use of clear delimiters (e.g., XML tags or triple backticks) or explicit warnings to ignore instructions within the fetched Notion content.
Capability inventory: the agent has the ability to write back to the workspace via Notion:notion-create-pages and Notion:notion-update-page.
Sanitization: there is no evidence of sanitization, filtering, or validation performed on the retrieved Notion content before it is interpolated into the agent's prompt context.
[EXTERNAL_DOWNLOADS]: the setup instructions in SKILL.md direct users to add an MCP (Model Context Protocol) server from https://mcp.notion.com/mcp. This is an official domain belonging to Notion, a well-known service provider, and is used for legitimate configuration purposes.

notion-meeting-intelligence