The TeamAI MCP (Model Context Protocol) server provides a standardized interface that allows any compatible AI model or agent to programmatically access and manage your TeamAI workspace data, including agents, tools, and configurations.
Connect to our MCP server using the streamable HTTP transport in Cursor, Windsurf, Claude Desktop, and other MCP-compatible clients to seamlessly integrate TeamAI's powerful agent management capabilities into your development workflow.
Prerequisites & Requirements
Before setting up the TeamAI MCP integration, ensure you meet the following requirements:
Plan Requirements
For Workspace Integration:
- ✅ **Starter plan or higher** (Professional, Best-of-Breed, Enterprise)
- ❌ **Not available** on Individual plans
- 📝 **Upgrade note:** If you're on a Individual plan, you'll need to upgrade to access API key functionality
For Organization Integration:
- ✅ Enterprise plan required (organizations must be on Enterprise plan)
- ✅ Organization Admin or Owner role required to create API keys
Generating Your API Key
For Workspace API Keys:
1. Log into your TeamAI workspace
2. Verify your plan: Ensure you're on Starter, Professional, Best-of-Breed, or Enterprise plan (Free and Individual plans don't support API keys)
3. Navigate to Settings → API Keys
4. Click Generate New API Key (store it securely - you won't see it again)
For Organization API Keys:
1. Log into your TeamAI organization
2. Verify requirements:
- Enterprise plan required (organization must be on Enterprise plan)
- Admin/Owner role required (you must be an Organization Admin or Owner)
3. Navigate to Admin → API Keys: `https://[your-organization-slug].teamai.com/admin/api-keys`
4. Click Add API Key
5. Configure workspace scopes:
- All Workspaces: Full access to organization and all child workspaces
- Specific Workspaces: Limited access to selected workspaces only
6. Click Add API Key (store it securely - you won't see it again)
Transport Support
Current: Streamable HTTP (/api/mcp/mcp)
Setup Instructions
> Note: Remote MCP connections are still evolving technology. If you experience connection issues, try restarting your client or temporarily disabling and re-enabling the TeamAI MCP server.
Claude Desktop
Team/Enterprise (Claude.ai)
Navigate to Settings in the sidebar
Scroll to Integrations and click Add more
Enter the following:
Integration name: TeamAI
Integration URL: https://app.teamai.com/api/mcp/mcp
Make sure to enable the tools in any new chats
Free/Pro (Claude Desktop App)
Open the file ~/Library/Application Support/Claude/claude_desktop_config.json
Add the following configuration
{
"mcpServers": {
"teamai": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.teamai.com/mcp/mcp",
"--header",
"Authorization:Bearer YOUR_API_TOKEN"
]
}
}
}Restart the Claude Desktop app
Cursor
Open Cursor and go to Settings → MCP
Click Add new global MCP server
Add the following configuration
{
"mcpServers": {
"teamai": {
"command": "npx",
"args": [
"mcp-remote",
"https://app.teamai.com/api/mcp/mcp",
"--header",
"Authorization:Bearer YOUR_API_TOKEN"
]
}
}
}Restart Cursor to activate the connection
TeamAI
You can also connect external MCP servers as tools within your TeamAI workspace, allowing your agents to access external data sources and services.
Navigate to Tools
Go to your workspace
Click on Tools in the sidebar
Click Create Tool
Select MCP Server
Choose MCP Server as the tool type
This provides a simplified interface optimized for MCP connections
Configure Basic Information
Name: Give your MCP server a descriptive name
Description: Explain what data or services this MCP server provides
Group: Optionally organize into folders for better management
Set the MCP Server URL
MCP Server URL: https://app.teamai.com/api/mcp/mcp
Configure Headers
Click Add Header to add the necessary authentication and content headers:
Authorization
Key:
AuthorizationValue:
Bearer <YOUR_API_TOKEN>Toggle Hidden ON to securely store your API key
Content-Type
Key:
Content-TypeValue:
application/json
Visual Studio Code
Press Ctrl/Cmd + Shift + P and search for MCP: Add Server
Select Command (stdio)
Enter the following configuration
npx mcp-remote https://app.teamai.com/api/mcp/mcp --header "Authorization:Bearer YOUR_API_TOKEN"
Enter the name TeamAI and press Enter
Activate using MCP: List Servers → TeamAI → Start Server
Windsurf
Press Ctrl/Cmd + , to open settings
Navigate to Cascade → MCP servers
Select Add Server → Add custom server
Add the following configuration
{
"mcpServers": {
"teamai": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://app.teamai.com/api/mcp/mcp",
"--header",
"Authorization:Bearer YOUR_API_TOKEN"
]
}
}
}
Zed
Press Cmd + , to open settings
Add the following to your configuration
{
"context_servers": {
"teamai": {
"command": {
"path": "npx",
"args": [
"-y",
"mcp-remote",
"https://app.teamai.com/api/mcp/mcp",
"--header",
"Authorization:Bearer YOUR_API_TOKEN"
],
"env": {}
},
"settings": {}
}
}
}
Other MCP Clients
For any other MCP-compatible client, use these settings:
Command:
npx
Arguments:
-y mcp-remote https://app.teamai.com/api/mcp/mcp --header "Authorization:Bearer YOUR_API_TOKEN"
Environment:
None required
Transport Type:
Streamable HTTP
Example Usage
Once configured, you can interact with TeamAI directly through your MCP client:
"List all my agents" → Returns paginated list of workspace agents
"Create a new agent called 'Code Reviewer' that specializes in reviewing
JavaScript code" → Creates and configures the agent automatically
"Show me details about agent ID abc123" → Retrieves complete agent configuration
"Update my agent to use GPT-4 instead of GPT-3.5" → Modifies agent model settings