All tutorials
Integrations & MCP
MCP (Model Context Protocol)
MCP (Model Context Protocol)
For advanced users, MCP lets you connect your Companions to virtually any external tool or service.
What MCP Is and Why It Matters
MCP (Model Context Protocol) is an open standard for connecting AI assistants to external tools and data sources. It is like a universal adapter that lets your Companion talk to any compatible service.
With MCP, you can:
- Connect to tools that do not have built-in Strawberry integrations
- Access internal company systems and databases
- Use custom tools built specifically for your workflow
- Extend your Companion's capabilities in powerful ways
How MCP Works
MCP lets Companions connect to MCP servers—small programs that expose tools and data to AI assistants.
When you connect an MCP server:
- Your Companion learns what tools are available
- You can ask your Companion to use those tools
- Your Companion calls the MCP server to perform actions
- Results come back to your conversation
Setting Up MCP Step by Step
- Find an MCP server — Look for MCP-compatible tools or create your own
- Get the connection details — You will need the server address and any authentication
- Open Settings → Integrations → MCP
- Add a new MCP connection — Enter the server details
- Test the connection — Your Companion will confirm what tools are available
- Start using it — Ask your Companion to use the new capabilities
When to Use MCP vs Built-In Apps
Use built-in apps when:
- There is already an integration for the service you need
- You want the simplest setup experience
- You do not need customization
Use MCP when:
- There is no built-in integration for your tool
- You need to connect to internal or custom systems
- You want more control over how integrations work
- You are building custom workflows
Requesting New Apps
If you would like to see a new integration or app supported in Strawberry, you can request it by emailing:
We review requests regularly and prioritize based on community demand. 🍓