How to work with Notion via MCP. Use when the user wants to read, search, create, or update Notion pages, databases, or workspace content.
Scanned 6/5/2026
Install via CLI
openskills install hi120ki/managed-agents-interface---
name: notion
description: How to work with Notion via MCP. Use when the user wants to read, search, create, or update Notion pages, databases, or workspace content.
---
## Finding and reading
Search by name or topic: `notion-search` → `notion-fetch`
Tasks / filtered queries: `notion-search` → `notion-query-database-view` or `notion-query-data-sources`
With Notion AI, `notion-search` also covers connected Slack / Drive / Jira
## Creating content
New page: ask where to save if unspecified, then `notion-create-pages`
Database entry: `notion-fetch` the DB first to get exact property names → `notion-create-pages`
New database: `notion-create-database` → `notion-create-view` for board/calendar/timeline
## Updating content
Property change: `notion-search` → `notion-update-page`
Targeted edit: `notion-fetch` → `notion-update-page` with `update_content`
Full rewrite: `notion-update-page` with `replace_content` — confirm first
Comment: `notion-create-comment`; read existing ones with `notion-get-comments`
DB schema change: `notion-update-data-source`
## Organizing
Move page: `notion-search` both ends → `notion-move-pages`
Duplicate: `notion-duplicate-page`
Add view: `notion-create-view`; edit view: `notion-update-view`
## Summarizing
Status report: `notion-search` → `notion-query-data-sources` → aggregate and present
Overdue tasks: query task DB with past due date + status not Done
By assignee: query task DB, group by person property
## People
Workspace members: `notion-get-users`; teamspaces: `notion-get-teams`
## Best Practices
- Search before fetch — semantic search usually resolves by name in one call
- Fetch DB schema before inserting — property names must match exactly
- Prefer `update_content` over `replace_content` to avoid overwriting others' work
- Ask where to save — pages without a parent go to private workspace
- Confirm before moving pages, bulk edits, or full rewrites
- Search rate limit is 30 req/min — avoid tight loops
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.