PyCharm is JetBrains' Python IDE, and MCP support is provided through the JetBrains AI Assistant plugin. PyCharm uses a project-level config at .ai/mcp/mcp.json — the same canonical mcpServers format as Claude Desktop, making it a passthrough generator. This file can be committed to version control to share server configurations with your team. The AI Assistant integrates MCP tools into its inline AI chat and code generation features. Because the format is identical to the canonical format, any server that works with Claude Desktop will work with PyCharm without any changes. Note that PyCharm must be fully closed and reopened for config changes to take effect — unlike some tools, it does not hot-reload the MCP config.
Quick Install with getmcp
The fastest way to install MCP servers in PyCharm is with the getmcp CLI:
npx @getmcp/cli add <server-id>
Configuration Format
Format
JSON
Root Key
mcpServers
Project Config
.ai/mcp/mcp.json
Prerequisites
•PyCharm (Community or Professional) installed (download from jetbrains.com/pycharm)
•JetBrains AI Assistant plugin installed (Plugins > Marketplace > AI Assistant)
•An active JetBrains AI subscription or trial
•Node.js 18 or higher for running npx-based servers
Popular Servers for PyCharm
Troubleshooting
▸Config changes not taking effect: PyCharm reads .ai/mcp/mcp.json only on startup. Fully close PyCharm (File > Exit) and reopen it after making changes — there is no hot-reload for MCP config.
▸AI Assistant plugin not installed: MCP support requires the JetBrains AI Assistant plugin. Go to Preferences > Plugins > Marketplace, search for AI Assistant, and install it. The plugin requires a JetBrains AI subscription.
▸Config directory not created: PyCharm does not create .ai/mcp/ automatically. Create the directories and file manually, or use the getmcp CLI which creates them if they do not exist.
▸MCP tools not appearing in AI chat: Ensure the AI Assistant plugin is enabled (not just installed) and that you are signed into a JetBrains account with an active AI subscription. Check Help > Diagnostic Tools > Show Log for plugin errors.
