The AI Tea
In everyday words
Notion is adding a way to run custom code and connect AI agents directly inside a Notion workspace, with a CLI to manage it from a terminal.
Need a meaning?
Notion’s hosted runtime for running custom code in a sandbox, used for integrations, sync logic, and deterministic agent tools.An API that lets third-party or in-house agents show up in Notion as native workspace participants that can chat and take actions.
Quick Sip
What you need to know
- Who is affected
- operations teams, developers, knowledge workers
- What changed
- On May 13, 2026, Notion announced its Developer Platform, including Workers (a hosted runtime for custom code), External Agents and an External Agent API (to bring third-party agents into Notion), and a new CLI called `ntn` for building, deploying, and managing these components.
- Why it matters
- Agent workflows often break at the “last mile”: you can chat with an AI, but you still need glue code, permissions, and reliable execution to connect tools to real work. Notion is trying to make that glue native to the workspace, so teams can deploy deterministic code and let agents act alongside humans inside Notion.
- What to watch next
- Watch how Worker permissions are scoped, what pricing and limits look like beyond the initial beta window, and how smoothly external agents can operate in busy, real-world Notion workspaces without creating security or audit gaps.
Four useful details
- Notion says Workers let developers deploy custom code to Notion’s hosted runtime in a sandbox, enabling sync jobs, webhooks, and custom agent tools.
- It announced External Agents plus an External Agent API intended to let third-party agents become first-class participants inside Notion workspaces.
- Notion introduced a CLI (`ntn`) for authentication, reading and writing to Notion, and deploying Workers from a terminal or IDE.
- The company emphasized built-in governance features like permissions and sandboxing from the start.
OpenAI · Official AnnouncementOffering Zero Data Retention for frontier models ↗
Adds source-backed context on ai safety from OpenAI.
OpenAI · Official UpdateAdvancing content provenance for a safer, more transparent AI ecosystem ↗Adds source-backed context on ai news from OpenAI.
Google DeepMind · Research Blog PostReimagining the mouse pointer for the AI era ↗Adds source-backed context on ai tools from Google DeepMind.
Your next sip
All latest briefings →