Code Ocean MCP Server

Search and run capsules, execute pipelines, and manage data assets on the Code Ocean platform.

Model Context Protocol (MCP) server for Code Ocean.

This MCP server provides tools to search and run capsules and pipelines, and manage data assets.

- Prerequisites
- Code Ocean Platform Version Compatibility
- Installation

- Visual Studio Code
- Claude Desktop
- Cline
- Roo Code
- Cursor
- Windsurf
- InstalluvfromAstralor theGitHub README
- Install Python 3.10 or newer usinguv python install 3.10(or a more recent version)
- Generate a Code Ocean access token. Follow instructions in theCode Ocean user guide.

Each release of this Code Ocean MCP Server is tested and verified against a specific minimum version of the Code Ocean platform API. Generally, this minimum version is the latest Code Ocean version at the time of the MCP Server release. We recommend ensuring your MCP Server dependency is pinned to a version compatible with your Code Ocean deployment. For details on when the minimum Code Ocean platform version changes, see theCHANGELOG.

{ ... "mcp": { "inputs": [ { "type": "promptString", "id": "codeocean-token", "description": "Code Ocean API Key", "password": true } ], "servers": { "codeocean": { "type": "stdio", "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "${input:codeocean-token}", "AGENT_ID": "VS Code" } } }, } }

- Open theclaude_desktop_config.jsonfile:

- macOS:~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:%APPDATA%\Claude\claude_desktop_config.json
- Under the top-level "mcpServers" object, add a "codeocean" entry. For a stdio transport (child-process) it looks like this:

{ "mcpServers": { "codeocean": { "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "<YOUR_API_KEY>", "AGENT_ID": "Claude Desktop" } } } }

Cline stores all of its MCP settings in a JSON file called cline_mcp_settings.json. You can edit this either through the GUI (“Configure MCP Servers” in the MCP Servers pane) or by hand:
- Open Cline and click the MCP Servers icon in the sidebar.
- In the “Installed” tab, click Configure MCP Servers → this opens your cline_mcp_settings.json.
- Add a "codeocean" server under the "mcpServers" key. For stdio transport:

{ "mcpServers": { "codeocean": { "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "<YOUR_API_KEY>", "AGENT_ID": "Cline" }, "alwaysAllow": [], // optional: list of tools to auto-approve "disabled": false // ensure it’s enabled } } }

- Save the file. Cline will automatically detect and launch the new server, making your Code Ocean tools available in chat .

Roo Code’s MCP support is configured globally across all workspaces via a JSON settings file or through its dedicated MCP Settings UI
- Click the MCP icon in Roo Code’s sidebar. 
- Select Edit MCP Settings (opens cline_mcp_settings.json). 
- Under "mcpServers", add:

{ "mcpServers": { "codeocean": { "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "<YOUR_API_KEY>", "AGENT_ID": "Roo Code" } } } }

- Save and restart Roo Code; your Code Ocean tools will appear automatically.

Optional: Manually editing cline_mcp_settings.json

Cursor stores MCP servers in a JSON file at either ~/.cursor/mcp.json (global) or {project}/.cursor/mcp.json (project-specific) .
- Open .cursor/mcp.json (or create it if missing). 
- Add under "mcpServers":

{ "mcpServers": { "codeocean": { "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "<YOUR_API_KEY>", "AGENT_ID": "Cursor" } } } }

- Save the file. Cursor will automatically detect and launch the new server on next start. 

Windsurf (Cascade) uses mcp_config.json under ~/.codeium/windsurf/ (or via the Cascade → MCP Servers UI) .
- Open your Windsurf Settings and navigate to Cascade → MCP Servers, then click View Raw Config to open mcp_config.json. 
- Insert the following under "mcpServers":

{ "mcpServers": { "codeocean": { "command": "uvx", "args": ["codeocean-mcp-server"], "env": { "CODEOCEAN_DOMAIN": "https://codeocean.acme.com", "CODEOCEAN_TOKEN": "<YOUR_API_KEY>", "AGENT_ID": "Windsurf" } } } }

- Save and restart Windsurf (or hit “Refresh” in the MCP panel).

You can test the MCP server locally during development withMCP Inspector:

npx @modelcontextprotocol/inspector uv tool run codeocean-mcp-server

The MCP server supports custom log formatting through theLOG_FORMATenvironment variable. This allows you to control the format of log messages output by the server.Example Format Strings:"%(asctime)s %(levelname)s [%(name)s] %(message)s". IfLOG_FORMATis not set, the server uses FastMCP's default logging configuration.

A Model Context Protocol (MCP) server that provides tools for AI, allowing it to interact with the DataWorks Open API through a standardized interface. This implementation is based on the Aliyun Open API and enables AI agents to perform cloud resources operations seamlessly.

An MCP server for interacting with Cloudera Machine Learning (CML).

Debug, build, and manage Microsoft Power Automate cloud flows with AI agents. 15 tools for action-level error details, flow creation, run history, and multi-tenant operations.

Manage your SiteBay WordPress hosting platform. Handle sites, execute server commands, and manage staging environments via natural language.

Integrate AI agents with StreamNative Cloud resources and Apache Kafka/Pulsar messaging systems.

Interact with your MLOps and LLMOps pipelines through your ZenML MCP server

Interact with Apache Airflow using natural language to manage and monitor your data workflows.

Provides tools to interact with the Ansible Automation Platform API for automation tasks.

Production-grade MCP server for Databricks: SQL Warehouses, Jobs API, multi-workspace support.