MCP servers for LM Studio
LocalLM Studio gives locally run open models MCP tools through mcp.json, editable from the Program tab in the chat sidebar.
How to add an MCP server to LM Studio
- In a chat, open the Program tab in the right sidebar and choose Install → Edit mcp.json.
- Add the server under "mcpServers".
- Load a model that supports tool calling and enable the server in the integrations list.
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/Users/you/projects"]
}
}
}
The example uses the reference filesystem server. Every listing on mcp.site shows its own command, args and environment variables, usually with a ready-made snippet for LM Studio.
Tool use quality depends on the model; pick one trained for function calling.
Config formats change between releases. If a step looks different in your version, the LM Studio documentation ↗ is the source of truth.
Servers to try in LM Studio
All 20,566 →Official GitHub MCP server for repository search, issues, pull requests, and CI/CD workflows in MCP-compatible AI clients.
Browser automation, page inspection, screenshots, and web interaction from Claude, Cursor, and other AI agents.
Up-to-date, version-specific library documentation and code examples inside AI coding prompts.
Powerful web scraping and search capabilities for LLM clients like Cursor and Claude.
Manage Supabase projects, databases, auth, storage, and SQL workflows from AI agents.
Deploy, configure & interrogate resources on the Cloudflare developer platform (Workers, KV, R2, D1).
Search engine made for AIs by Exa.
Connect AI agents to Notion pages, databases, and workspaces.
Query orders, customers, and products from your WooCommerce store.
Create, edit, and schedule Ghost blog posts and newsletters.
FHIR-native healthcare data platform tools for AI agents.
Persistent long-term memory graph shared across MCP clients.