Enterprise DNA

Omni by Enterprise DNA

Enterprise DNA Resources

Latest AI and industry news. Practical AI operating-system thinking for owners, operators, and teams doing real work.

220k+

Data professionals

Omni

AI agents and apps

Audit

Map the manual work

News Trending AI News

Notion Turns Its Workspace Into an AI Agent Control Room

Notion's new Developer Platform with Workers, External Agents API, and CLI lets teams run AI agents directly inside their workspace.

Enterprise DNA | | via TechCrunch
Enterprise DNA News

Notion just made its biggest bet on AI agents yet. On May 13, 2026, the company launched its Developer Platform, introducing a new primitive called Workers, an External Agents API, and a command-line interface that turns Notion from a knowledge workspace into a coordination layer for AI agents.

The launch comes roughly three months after Notion released Custom Agents in February 2026. Since then, customers have built over one million agents inside the platform — a number that signals genuine adoption, not just experimentation.

What Notion Built

The centerpiece of the Developer Platform is Workers — a hosted runtime that lets teams write and deploy custom code directly inside Notion. Workers run in a secure sandbox, powered by Vercel’s infrastructure, meaning teams don’t need to manage their own servers. The code runs inside Notion, alongside the data it is meant to act on.

Workers enable three things teams have been piecing together with external tools:

Database sync — pull any external data source via API directly into a Notion database, with no infrastructure to maintain.

Webhook support — any external app can now trigger Notion directly. A Worker receives the event, runs logic, and takes action in Notion or calls other services.

Automated multi-step workflows — sequences that combine data, decisions, and actions, all running inside the workspace.

The External Agents API

Alongside Workers, Notion launched the External Agents API — a way to bring AI agents from outside Notion into the workspace as first-class collaborators. Agents appear on the same surfaces as human teammates, with their own permissions and tool access.

Notion has partnered with Claude, Codex, Cursor, and Decagon as out-of-the-box integrations. Teams can also connect agents they built themselves through the same API.

This matters because instead of routing work through a separate agent orchestration platform, teams can coordinate AI agents from within their existing knowledge workspace. The agents work where the data already lives.

The CLI

Notion also released a developer command-line interface, available on all plans. It lets developers — and coding agents — sign in, read and write Notion content, build and deploy Workers, and extend Notion’s behavior for their team’s specific context.

Workers deployment and management is available on Business and Enterprise plans. Workers are free to use during the beta period.

Why Notion Is Building This Now

The timing is deliberate. As AI agents become more capable and more common in business workflows, the question of where they run and what data they can access has become central. Every major enterprise software platform is now racing to become the layer that coordinates AI work.

Salesforce has Agentforce. ServiceNow has its agent ecosystem. Microsoft is building Agent 365. Notion’s answer is different: build the agent infrastructure directly into the knowledge workspace, where your team’s processes and information already live.

For many small and mid-size businesses, Notion is already where documentation, projects, wikis, and databases live. The Developer Platform is an attempt to make Notion the place where AI agents run too, rather than having teams buy a separate orchestration tool.

What This Means for Business

The appeal of the Notion approach is consolidation. Businesses that have been building AI workflows across disconnected tools — an agent platform here, a database there, webhooks through another service — now have a path to bring that inside one workspace.

Consider a practical example: a business could write a Worker that syncs CRM data into a Notion database, connects to a Claude agent that reviews and summarizes new leads, and triggers a workflow that notifies the right team member and creates a task automatically. No external infrastructure. No separate subscriptions to stitch together. It runs inside Notion.

For teams evaluating AI agent strategy, the practical question has shifted. It’s no longer just which AI model to use. It’s also where AI agents should live — and whether you want them operating in a separate system or embedded in the workspace where your team actually works.

Notion’s Developer Platform is a genuine step forward in making AI agents accessible to businesses that don’t have engineering teams dedicated to AI infrastructure. For many businesses, that accessibility will matter more than raw capability.


If you’re working out how AI agents fit into your business operations — where they run, what they connect to, and how to make them useful day to day — that’s exactly what Enterprise DNA’s Omni Advisory service is built for. Book a discovery call to talk through your situation.