switchboard
G

Google Colab MCP Server

by Google

Official MCP server for Google Colab. Lets AI agents create, execute, and manage GPU-accelerated Jupyter notebooks via Model Context Protocol — no browser needed.

5
Skills
OAuth 2.0
Auth
Yes
Streaming
No
Push

Skills

Create and manage Jupyter notebooks programmatically from any MCP-compatible AI agent or coding assistant

Create and manage Jupyter notebooks programmatically from any MCP-compatible AI agent or coding assistant

Execute Python code cells in Google's GPU/TPU-accelerated cloud runtime without browser interaction

Execute Python code cells in Google's GPU/TPU-accelerated cloud runtime without browser interaction

Install and manage Python package dependencies dynamically within a running Colab session

Install and manage Python package dependencies dynamically within a running Colab session

Run end-to-end data analysis pipelines and return structured results back to the orchestrating agent

Run end-to-end data analysis pipelines and return structured results back to the orchestrating agent

Automate multi-step ML experiments and chain notebook outputs into downstream workflow automation

Automate multi-step ML experiments and chain notebook outputs into downstream workflow automation

Code & DevToolsnotebook-automationcloud-computepython-executiondata-analysisjupytergpu-accessmcp-protocolopen-source
Visit Agent

Related Agents

O

OpenSandbox

Runs AI-agent workloads in isolated Docker or Kubernetes sandboxes, exposing sandbox lifecycle, command, filesystem, an…

Infrastructure & OpsCode & DevTools
APIMCPCLI
5 skillsAPI Key
B

BitFun

Runs an open-source desktop and terminal coding agent that plans, edits, tests, and reviews changes inside real Git rep…

Code & DevToolsBrowser & Computer Use
CLI
3 skillsAPI Key
P

Proliferate

Open-source, self-hostable coding agent platform for running cloud coding sessions on your own infrastructure.

Code & DevToolsInfrastructure & Ops
CLIAPI
0 skillsAPI Key
S

SandBase Harness

Runs local-first, self-hosted AI agents with persistent sessions, sandboxed tools, memory, credentials, audit trails, r…

OrchestrationCode & DevTools
CLIMCP
3 skillsNone
google-colab-mcp
Official MCP server for Google Colab. Lets AI agents create, execute, and manage GPU-accelerated Jupyter notebooks via Model Context Protocol — no browser needed.
fields
nameGoogle Colab MCP Server
providerGoogle
urlhttps://github.com/googlecolab/colab-mcp
categoriescode-devtools
accessmcp · api
authoauth2
streamingtrue
pushfalse
verifiedtrue
tagsnotebook-automation, cloud-compute, python-execution, data-analysis, jupyter, gpu-access, mcp-protocol, open-source
skills
create-and-manage-jupyter-notebooks-programmatically-from-any-mcp-compatible-ai-agent-or-coding-assistantCreate and manage Jupyter notebooks programmatically from any MCP-compatible AI agent or coding assistantCreate and manage Jupyter notebooks programmatically fr…
execute-python-code-cells-in-google-s-gpu-tpu-accelerated-cloud-runtime-without-browser-interactionExecute Python code cells in Google's GPU/TPU-accelerated cloud runtime without browser interactionExecute Python code cells in Google's GPU/TPU-accelerat…
install-and-manage-python-package-dependencies-dynamically-within-a-running-colab-sessionInstall and manage Python package dependencies dynamically within a running Colab sessionInstall and manage Python package dependencies dynamica…
run-end-to-end-data-analysis-pipelines-and-return-structured-results-back-to-the-orchestrating-agentRun end-to-end data analysis pipelines and return structured results back to the orchestrating agentRun end-to-end data analysis pipelines and return struc…
automate-multi-step-ml-experiments-and-chain-notebook-outputs-into-downstream-workflow-automationAutomate multi-step ML experiments and chain notebook outputs into downstream workflow automationAutomate multi-step ML experiments and chain notebook o…
→ https://github.com/googlecolab/colab-mcp