Skip to main content

MCP Servers

Manage your Model Context Protocol servers and their available tools.

What is MCP?

The Model Context Protocol (MCP) is an open protocol that enables AI assistants to connect to external data sources and tools. MCP provides a standardized way for AI applications to access context from various systems.

Overview

The MCP Servers page displays all servers available in your workspace. Each server provides a set of tools that your AI assistant can use.

Server List

Each server card shows:

  • Server name and icon - Identifies the service (e.g., GitHub, Figma, Jira)
  • Connection status - Green checkmark indicates connected
  • Available tools - List of tools provided by the server (shown as tags)
  • Use toggle - Enable/disable the server for your AI client
  • Use button - Quick action button

Example Servers

ServerTools
Context7resolve-library-id, get-library-docs
Figmaget-file, get-image, get-file-versions, get-team-projects
GitHubcreate_or_update_file, search_repositories, create_repository, push_files, create_pull_request, search_code, search_issues, merge_pull_request, get_pull_request_files
Jiraget-issues, list-fields, list-issue-types, list-link-types, get-user, create-issue, create-issue-link, get-create-meta-data, comment-on-issue, get-issue-changelog

Searching Servers

Use the search bar at the top of the page to filter servers by name.

Connecting to a Server

Servers that are not yet connected show a Connect button. Click it to:

  1. Start the OAuth flow (if required)
  2. Grant permissions to the target service
  3. Return to Webrix with the connection established

Once connected, the server shows a green checkmark next to its name.

Using Servers

Toggle the Use switch to enable or disable a server for your AI client. When enabled:

  • The server's tools become available to your AI assistant
  • You can use the tools through the MCP connection

Server Menu

Click the three-dot menu (⋮) on a server card for additional options.