Skip to main content

Documentation Index

Fetch the complete documentation index at: https://anaconda.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

anaconda-mcp is a package that exposes conda environment management tools to MCP-enabled AI coding assistants using the anaconda mcp commands. It acts as a unified MCP endpoint, giving AI assistants like Claude, Cursor, and VS Code awareness of your conda environments, packages, and channel configurations. This guide provides an overview of the anaconda mcp commands and core concepts. For a walkthrough of how to connect Anaconda MCP to your AI coding assistant automatically using a in Anaconda Desktop, see Anaconda MCP.

​
Installation

conda install anaconda-mcp
Anaconda recommends installing anaconda-mcp in its own dedicated conda environment rather than base to avoid Python version and dependency conflicts.

​
Basic usage

​
Configure your AI coding assistant

This command launches an interactive wizard to configure your AI coding assistant to use Anaconda MCP Server at a global or project level. Anaconda MCP currently supports the following assistants: Claude Desktop, Claude Code, Cursor, VS Code, Opencode, and Windsurf. You must be logged in to your anaconda.com account to use Anaconda MCP. If you’re not logged in, anaconda mcp setup prompts you to log in.
You are prompted to agree to the Anaconda MCP Beta Agreement during initial setup.
anaconda mcp setup
To configure a specific assistant non-interactively, use the --client flag: 
anaconda mcp setup --client <CLIENT>
The following client identifiers are accepted by the --client flag:

​
Remove a configuration

This command removes the Anaconda MCP Server configuration for an AI coding assistant. --client is required. 
anaconda mcp remove --client <CLIENT>

​
List supported assistants

This command lists all supported AI coding assistants and shows whether Anaconda MCP Server is currently configured for each. 
anaconda mcp clients

​
Capabilities

Once connected, you can manage your conda environments and packages using natural language: