Step-by-step guides to install and configure MCP servers in 19 AI applications. Each guide covers the config format, file location, quick install, and troubleshooting.
Claude Desktop is Anthropic's flagship desktop application for interacting with Claude. It was the first major AI app to support the Model Context Protocol, making it the reference implementation for MCP configuration.
~/Library/Application Support/Claude/claude_desktop_config.json
VS Code added MCP support through GitHub Copilot Chat, allowing Copilot to call tools from MCP servers during conversations. The config file lives at .
.vscode/mcp.json
Cursor is an AI-powered code editor built on VS Code with deep AI integration. It supports MCP servers using the same JSON format as Claude Desktop — an mcpServers root key with stdio and remote server configs — making it one of the easiest apps to migrate between.
.cursor/mcp.json
Windsurf is an AI-first code editor developed by Codeium, featuring the Cascade AI agent with deep agentic capabilities. It supports MCP servers through a global config file at ~/.
~/.codeium/windsurf/mcp_config.json
Goose is Block's open-source autonomous AI coding agent that runs in the terminal and can take multi-step actions. Its MCP support is unique among popular AI apps: it uses YAML format instead of JSON, and has its own field naming conventions.
~/.config/goose/config.yaml
Claude Code is Anthropic's CLI-based agentic coding tool that runs in the terminal and can autonomously read, write, and execute code. It supports MCP servers through a format nearly identical to Claude Desktop — an mcpServers root key with the same stdio and remote server structure.
.mcp.json
Cline is a popular AI coding assistant VS Code extension that integrates Claude and other models directly into your editor. It stores its MCP server config in VS Code's global extension storage rather than a workspace file, meaning servers are available across all your projects.
~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
Roo Code (formerly Roo Cline) is an AI coding assistant VS Code extension that evolved from the Cline codebase with additional features for autonomous task execution. Its MCP config format is very similar to Cline: it uses the mcpServers root key and adds alwaysAllow and disabled to each server entry.
~/Library/Application Support/Code/User/globalStorage/rooveterinaryinc.roo-cline/settings/mcp_settings.json
OpenCode is an open-source AI coding agent by Anomaly that runs in the terminal with support for multiple AI models. Its MCP configuration differs from the canonical format in several meaningful ways: the root key is mcp instead of mcpServers, stdio servers use an array-format command that merges the command binary and all its arguments into a single list, and environment variables go in an environment field instead of env.
opencode.json
Zed is a high-performance code editor built in Rust by Zed Industries, designed for speed and collaborative editing. Zed added MCP support through its context servers feature, which allows the AI assistant to call external tools during conversations.
~/.config/zed/settings.json
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
Codex is OpenAI's AI coding agent that runs in the terminal, designed for autonomous software engineering tasks. It is the only major AI app in the getmcp ecosystem that uses TOML for its config format.
.codex/config.toml
Gemini CLI is Google's open-source AI agent for the terminal, powered by the Gemini model family. It reads MCP server configuration from ~/.
~/.gemini/settings.json
Continue is a popular open-source AI code assistant that integrates as an extension into VS Code and JetBrains IDEs. It supports MCP servers through a global config file at ~/.
~/.continue/config.json
Amazon Q Developer is AWS's AI coding assistant, available as a CLI tool and as IDE extensions for VS Code and JetBrains. The CLI variant reads MCP server configuration from ~/.
~/.aws/amazonq/mcp.json
Trae is an AI-powered IDE developed by ByteDance that brings agentic coding capabilities to a VS Code-based environment. It uses project-scoped MCP configuration stored at .
.trae/mcp.json
BoltAI is a native macOS AI chat application that provides a system-wide AI assistant integrating with multiple model providers. It supports MCP servers through a global config file at ~/Library/Application Support/BoltAI/mcp_config.
~/Library/Application Support/BoltAI/mcp_config.json
LibreChat is an open-source, self-hosted web application that provides a ChatGPT-like interface supporting multiple AI providers including OpenAI, Anthropic, Google, and more. It supports MCP servers through a YAML configuration section in the root librechat.
librechat.yaml
Antigravity is Google's AI-first IDE, designed for agentic software development with deep integration of Gemini models. It stores MCP server configuration at ~/.
~/.gemini/antigravity/mcp_config.json