{"categories":[{"label":"Application Frameworks","url":"https://skillfed.io/packages/category/software-development-libraries-application-frameworks/5"}],"enrichment":{"capability":"An MCP server that exposes Jupyter notebooks and code execution to AI agents and clients, enabling real-time notebook management, cell execution, and sandbox-backed code runs across local and cloud environments.","skillfed_tags":["jupyter-integration","ai-agent-tools","mcp-protocol"],"use_cases":["Enable AI agents to execute code in local or cloud Jupyter notebooks with real-time feedback and error recovery.","Route code execution across multiple sandbox backends without changing client code.","Build multi-notebook workflows where an agent switches between notebooks and manages kernel state automatically.","Integrate Jupyter execution into an MCP-compatible agent framework for persistent, context-aware data analysis.","Use the hosted Datalayer MCP endpoint to connect any MCP client to notebooks without running a local server."],"what_it_does":"Jupyter MCP Server is a Model Context Protocol server that bridges Jupyter notebooks and AI agents, allowing tools to read, write, and execute code in notebooks in real-time. It exposes MCP tools for notebook management (switching between notebooks, listing kernels), cell operations (reading and executing cells, managing outputs), and code sandbox execution across multiple backends including local Jupyter, Datalayer, Kaggle, Google Colab, and Modal.\n\nThe package is built on fastapi and jupyter-server, with observability via opentelemetry-api and opentelemetry-sdk. It can run as a standalone MCP server, as a Jupyter extension, or connect to existing Jupyter deployments. Datalayer hosts a public instance for immediate use without local setup. The primary use case is enabling AI agents to work with notebooks persistently\u2014executing code, handling failures, and switching execution backends on demand.","worth_installing":"Yes, if you use Jupyter with AI agents or need to expose notebook execution via MCP. The package is actively maintained, permissively licensed, has low install friction, and solves a specific gap: real-time agent control of notebooks. Install it locally to connect your Jupyter server, or use the hosted endpoint for immediate integration. No known vulnerabilities as of 2026-08-14."},"id":"jupyter-mcp-server","links":{"html":"https://skillfed.io/packages/jupyter-mcp-server","md":"https://skillfed.io/packages/jupyter-mcp-server.md","pypi":"https://pypi.org/project/jupyter-mcp-server/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-14","license_spdx":null,"license_treatment":"permissive","name":"jupyter-mcp-server","python_support":"supports_current","summary":null},"popularity":{"monthly_downloads":92586,"position":13441,"tier":"top_15000"},"security":{"n_vulnerabilities":0},"version":"1.3.8"}