Jupyter AI

Frontier AI Agents via ACP

Jupyter AI integrates with frontier AI agents including Claude, Codex, GitHub Copilot, Gemini, Goose, Kiro, Mistral Vibe, and OpenCode through the Agent Client Protocol (ACP). Agents are automatically detected when their dependencies are installed.

Collaborate with a wide range of leading AI agents directly in JupyterLab without manual configuration.

Built-in Jupyter MCP Server

Agents can read and write files, run terminal commands, and interact with notebooks through a built-in Jupyter MCP server, enabling seamless computational workflows.

Automate notebook tasks and file operations with AI agents while keeping everything inside your Jupyter environment.

Guardrails for Agent Actions

A permission system gives you guardrails over agent actions — agents request approval before writing files or executing commands, ensuring safe and controlled operations.

Maintain control and security when AI agents perform actions on your behalf.

Native Chat Interface

Jupyter AI provides a native chat UI where you can create multiple concurrent chats, drag and drop files or notebook cells as context, and collaborate in real time with other users connected to the same server.

Streamline your AI interactions with a flexible, multi-threaded chat experience.

Custom MCP Servers & Personas

You can add custom MCP servers to give agents access to domain-specific tools, resources, and prompts. Developers can build and register their own AI personas using the entry points API.

Tailor Jupyter AI to your specific workflows and domain needs.

Built on ACP and MCP

By building on open standards like ACP and MCP, Jupyter AI avoids vendor lock-in and gives you access to the full ecosystem of compatible agents and tools.

Future-proof your AI setup with interoperable, standards-based technology.

• Open Source and Free to UseJupyter AI is an open source extension that connects AI agents to computational notebooks in JupyterLab. It is licensed under the BSD-3-Clause license, making it freely available for use and modification.
• Supports Multiple AI AgentsJupyter AI provides a native chat UI where you can collaborate with frontier AI agents including Claude, Codex, GitHub Copilot, Gemini, Goose, Kiro, Mistral Vibe, and OpenCode. Agents are automatically detected when their dependencies are installed.
• Built on Open StandardsJupyter AI is built on open standards like the Agent Client Protocol (ACP) and Model Context Protocol (MCP). This avoids vendor lock-in and gives access to the full ecosystem of compatible agents and tools.
• Extensible and CustomizableJupyter AI is designed to be flexible and extensible. You can add custom MCP servers to give agents access to domain-specific tools, resources, and prompts. Developers can build and register their own AI personas using the entry points API.

• Incubation StatusJupyter AI is currently under incubation as part of the JupyterLab organization. This means it is still in an early stage of development and may not be as stable or mature as fully established projects.
• Requires Agent DependenciesTo use a specific AI agent, you must install the agent's dependencies separately. Agents are only automatically detected when their dependencies are installed, so you need to ensure the correct packages are set up.
• Permission System for ActionsJupyter AI includes a permission system that gives guardrails over agent actions. Agents request approval before writing files or executing commands, which may require user interaction and could slow down workflows.
• Potential for ErrorsAs with any AI-powered tool, there is a risk of errors or unexpected behavior. The GitHub repository shows 286 open issues, indicating that users may encounter bugs or limitations that need to be addressed.

Collaborate with Frontier AI Agents

Jupyter AI provides a native chat UI in JupyterLab where you can collaborate with frontier AI agents including Claude, Codex, GitHub Copilot, Gemini, Goose, Kiro, Mistral Vibe, and OpenCode, all integrated through the Agent Client Protocol (ACP). Agents are automatically detected when their dependencies are installed, making setup as simple as installing Jupyter AI and the agent of your choice.

Automate Notebook and File Operations

Agents in Jupyter AI can read and write files, run terminal commands, and interact with notebooks through a built-in Jupyter MCP server. This enables you to delegate repetitive coding and data tasks directly from your notebook environment, streamlining your workflow.

Control Agent Actions with Permissions

A permission system gives you guardrails over agent actions — agents request approval before writing files or executing commands. This ensures you maintain control and security while leveraging AI assistance in your computational notebooks.

Extend with Custom MCP Servers and Personas

Jupyter AI is designed to be flexible and extensible. You can add custom MCP servers to give agents access to domain-specific tools, resources, and prompts. Developers can also build and register their own AI personas using the entry points API, avoiding vendor lock-in and accessing the full ecosystem of compatible agents and tools.

Define the exact google Colab AI Tools workflow Jupyter AI should support.

Compare it with closely related AI tools in the same category before committing.

Set review rules for accuracy, privacy, brand voice, compliance, and final approval.

Connect useful outputs to the wider stack instead of leaving them inside the AI tool.