Read and write Notion databases, pages, blocks and users via the Notion API.
Notion ships in the w6w first-party pack. It declares 17 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.notionNotion reads and writes pages, databases and blocks from a workflow — creating a page directly or as a new row in a database, updating a database page’s properties, and appending blocks to build out a page’s content programmatically.
Databases are a first-class part of the surface: they can be listed, searched and read individually, and their pages fetched one at a time or in bulk, which is what lets a workflow treat a Notion database as a lightweight structured store rather than only writing unstructured pages. Pages can be archived and restored, and both pages and databases are searchable by title across a workspace. Users can be listed or looked up individually too.
A fit for turning external data into a new Notion page or database row, keeping a database in sync with another system, or pulling structured content out of Notion for use elsewhere in a workflow.
Three routes to the same 17 actions. The Workflow tab is generated from Notion's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
create-database-page Create a page inside a database. Provide the `properties` map keyed by property name.
get-many-blocks List the direct children of a block (or a page). Returns one page — pass `startCursor` back to walk.
get-many-database-pages Query the pages in a database. Returns one page — pass `startCursor` back to walk.
get-many-databases Walk /search until every database has been fetched (or `limit` is reached).
get-many-users List every user in the workspace. Returns one page — pass `startCursor` back to walk.
list-databases List all databases the integration has access to. One page per call — pass `startCursor` back to walk.
search-databases Search databases by title. Returns one page — pass `startCursor` back to walk.
update-database-page Update a database page's properties, icon, cover, or archived flag.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Notion's real ids.
{
"manifestVersion": "2",
"name": "notion-example",
"steps": [
{
"id": "append-block-children",
"uses": {
"app": "io.w6w.notion",
"action": "append-block-children",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"blockId": "<blockId>",
"children": "<children>"
}
}
]
}append-block-children create-database-page create-page get-block get-database +12 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Notion, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Notion. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "create-page",
payload: {
parentPageId: "<value>",
title: "<value>",
// children: "<value>",
// icon: "<value>",
// cover: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action create-page --payload '{"parentPageId":"<value>","title":"<value>"}' Give an AI agent Notion — without giving it Notion's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.notion#append-block-children",
"input": {
"blockId": "<blockId>",
"children": "<children>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Notion connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Notion's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Notion itself is MIT, and the runtime that executes it is source-available (FSL).
Notion declares its own checks, so its health is a property of the app rather than something the host guesses at.