MCP (Model Context Protocol) servers extend Projects capabilities by connecting it to external tools and services. This allows the AI to access and interact with your existing business systems, from project management tools to documentation platforms.
MCP servers give the AI model awareness of external tools and services it can access. Instead of being limited to its built-in capabilities, Projects can integrate with your existing workflow tools through these protocol connections.
For details about the Model Context Protocol, visit the official MCP documentation.
Projects includes ready-to-use integrations with popular enterprise tools.
To connect a built-in integration:
- Go to MCP Servers.
- Click + Connect on the built-in integration tile for your chosen service and provide the requested data when prompted.
The video below shows where to connect one of the Built-in integrations, in this case, Atlassian:
Connect to Jira and Confluence through OAuth authentication. Once connected, Projects can:
- Access Jira tickets and Confluence
- Automatically fix issues
More built-in integrations, such as Google Drive, Supabase, and Zapier are in active development.
Enterprise customers can build and deploy their own MCP servers using the remote Model Context Protocol standard. This provides a way to:
- Connect Projects to proprietary internal tools
- Integrate with custom APIs and databases
- Extend Projects capabilities to match your specific workflow needs
To connect your custom MCP Server:
- Go to MCP Servers.
- In the Your Custom Servers section, click + Add Your First Server.
- In the dialogue that opens, provide the Server Name, the Server URL, and the optional Authorization Token.
The video below shows where to connect your custom server:
After you have configured your MCP servers, they are available within your Projects workspace. To display all connected MCP servers in the Projects interface:
- Open one of your Projects on the Projects page.
- On the Generate tab, click on the MCP Servers icon to open the MCP Server dialogue.
Once connected, the AI automatically leverages these services when relevant to your tasks, with MCP servers enhancing Projects capabilities in the background.
For example, when working on a development task, Projects can automatically access related Jira tickets, understand the requirements, and even implement fixes based on the ticket context.
The video below shows where to find the MCP Servers in the Projects interface with an open project:
For more information on working with Builder Projects and Builder's AI, visit:
- Builder Figma Plugin. Learn how to use the Builder Figma plugin to get your design into Builder.
- AI Credits. Understand Builder's AI credits and how and when they apply.
- Builder Projects: best practices. Get tips and insights to work even smarter.
- Connect GitHub to Projects. Connect your repo to Builder so you can edit your code visually and create PRs with Builder.