Productivity software company Notion announced a strategic shift on Wednesday, unveiling a new Developer Platform designed to transform its workspace into a central hub for AI agents. This move marks Notion's evolution from a collaborative note-taking application into a programmable infrastructure capable of coordinating complex automated workflows. The new platform introduces an orchestration layer that connects Notion's custom AI agents with external tools and databases, enabling teams to execute multi-step processes without relying on third-party automation scripts. Previously, Notion launched Custom Agents in February to handle repetitive tasks such as answering FAQs and compiling status updates. While over one million such agents were created by customers, the system faced significant limitations. These agents could not access external data or execute custom logic, and external AI agents used by companies had no direct connection to the Notion ecosystem. Users were forced to build workarounds using external infrastructure or custom scripts, a gap that co-founder and CEO Ivan Zhao acknowledged during the live announcement. The introduction of a feature called Workers addresses these constraints by providing a cloud-based environment for running custom code. Users can deploy logic into a secure sandbox, allowing them to sync data, build custom tools, and trigger actions via webhooks directly within Notion. Notably, users do not need to write code manually, as AI coding assistants can generate the necessary scripts. To encourage adoption, Workers will be free to use through August, utilizing the same credit system as Custom Agents. Data integration is further enhanced through database sync capabilities powered by Workers. This feature allows Notion to pull real-time data from any database with an API, including major platforms like Salesforce, Zendesk, and Postgres. This ensures that Notion databases can serve as a dynamic canvas for both workflows and agents, keeping information current across the organization. The platform also supports the Model Context Protocol (MCP), an emerging standard for connecting AI tools to external services. When standard connections are insufficient, Workers enable the creation of agent tools with custom logic. Furthermore, Notion now allows users to chat directly with external AI agents, assigning them tasks and tracking progress as if they were internal employees. Supported partner agents at launch include Claude Code, Cursor, Codex, and Decagon, with an API available for companies to integrate their own internal systems. Developers interact with the new ecosystem via the Notion CLI, a command-line tool available on Business and Enterprise plans. This launch represents a broader industry trend where AI companies are moving beyond simple chatbots toward agentic tools that can perform actions across multiple software platforms. By combining agents, custom code, and live data in a single location, Notion aims to become core infrastructure for knowledge work automation. Zhao summarized the vision of the new platform by stating that it now enables the use of any data, tool, or agent within a unified environment.