Step-by-step instructions for connecting Marker.io MCP to your AI tool. For an overview of what MCP can do, see MCP Integration (Model Context Protocol).
The MCP server URL is:
https://connect.marker.io/mcp
Authentication
There are two ways to authenticate:
OAuth - your AI tool opens a browser window where you approve access. No token needed.
Access token - for clients that don't support OAuth, generate a bearer token from your Marker.io account settings.
ChatGPT
ChatGPT supports MCP via Connectors (requires ChatGPT Plus/Pro or Teams).
Go to Settings > Apps & Connectors > Advanced settings and enable Developer mode
Go to Settings > Connectors > Create
Enter name:
Marker.ioEnter MCP server URL:
https://connect.marker.io/mcpSet authentication to OAuth
Click Create, then complete the OAuth login flow
Codex
Open your terminal and run:
codex mcp add --url https://connect.marker.io/mcp marker-io
Then authenticate:
codex mcp login marker-io
Complete the authorization flow in your browser.
Claude.ai
Open Claude.ai → Settings → Connectors
Click Add custom connector
Enter the MCP server URL:
https://connect.marker.io/mcpClick Add and complete the authorization flow
Claude Code
claude mcp add --transport http marker-io https://connect.marker.io/mcp
Claude Desktop
Edit the config file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
{ "mcpServers": { "marker-io": { "command": "npx", "args": [ "mcp-remote", "https://connect.marker.io/mcp", "--header", "Authorization: Bearer YOUR_TOKEN" ] } }}
Replace YOUR_TOKEN with your access token. Restart Claude Desktop after editing.
Cursor
Create or edit .cursor/mcp.json in your project root (or ~/.cursor/mcp.json for global setup):
{ "mcpServers": { "marker-io": { "url": "https://connect.marker.io/mcp", "headers": { "Authorization": "Bearer YOUR_TOKEN" } } }}
Replace YOUR_TOKEN with your access token. Cursor will handle the OAuth flow automatically if you omit the headers.
VS Code
Create .vscode/mcp.json in your workspace, or use the Command Palette > "MCP: Add Server":
{ "servers": { "marker-io": { "type": "http", "url": "https://connect.marker.io/mcp" } }}
Requires VS Code 1.101+.
GitHub Copilot
Open VS Code → Settings → search for MCP. Click Edit in settings.json and add:
{ "servers": { "marker-io": { "type": "http", "url": "https://connect.marker.io/mcp" } }}
Make sure Agent Mode is enabled in GitHub Copilot Chat.
Windsurf
Edit ~/.codeium/windsurf/mcp_config.json:
{ "mcpServers": { "marker-io": { "serverUrl": "https://connect.marker.io/mcp" } }}
Windsurf will open an OAuth authorization flow to connect. Restart Windsurf after editing.
JetBrains
Open Settings → Tools → AI Assistant → Model Context Protocol (MCP)
Click Add to add a new MCP server
Paste the following configuration:
{ "mcpServers": { "marker-io": { "url": "https://connect.marker.io/mcp", "headers": { "Authorization": "Bearer YOUR_TOKEN" } } }}
Replace YOUR_TOKEN with your access token.
OpenCode
Edit ~/.config/opencode/opencode.json (or opencode.json in your project root):
{ "mcp": { "marker-io": { "type": "remote", "url": "https://connect.marker.io/mcp", "enabled": true } }}
CMS-specific guides
If your team uses a CMS, we have dedicated guides for connecting Marker.io MCP to your platform:
Other clients
For any MCP-compatible client not listed above:
Set the server URL to
https://connect.marker.io/mcpAdd the
Authorization: Bearer YOUR_TOKENheader with your access token
{ "mcpServers": { "marker-io": { "url": "https://connect.marker.io/mcp", "headers": { "Authorization": "Bearer YOUR_TOKEN" } } }}
Replace YOUR_TOKEN with your access token.
Need Help?
If you have any questions, comments, or corrections, chat with us at the bottom right of our web pages.
