Use this file to discover all available pages before exploring further.
SegmentStream MCP is a self-contained server with built-in measurement skills. Connect your AI tool using the server URL — authentication is handled automatically via OAuth.
The fastest way to get started. Once connected in claude.ai, the connection is automatically available in Claude Code, Claude Desktop, and Cowork — no additional setup needed.
1
Click Customize
Go to claude.ai and click Customize in the left sidebar.
2
Open Connectors
Click Connectors.
3
Add a custom connector
Click the + button, then select Add custom connector.
4
Enter the SegmentStream MCP URL
Enter a name (e.g. “SegmentStream”) and paste the server URL, then click Add:
https://mcp.segmentstream.com/mcp
5
Connect
The connector appears as “Not connected”. Click Connect to start the OAuth flow.
6
Authorize
Sign in with your SegmentStream account and click Allow.
Cowork uses the same Connectors interface as claude.ai.
1
Click Customize
Click Customize in the left sidebar.
2
Open Connectors
Click Connectors.
3
Add a custom connector
Click the + button, then select Add custom connector.
4
Enter the SegmentStream MCP URL
Enter a name (e.g. “SegmentStream”) and paste the server URL, then click Add:
https://mcp.segmentstream.com/mcp
5
Connect
The connector appears as “Not connected”. Click Connect to start the OAuth flow.
6
Authorize
Sign in with your SegmentStream account and click Allow.
Once connected, start a new task and ask the agent about your marketing data.
Run this command in your terminal:
claude mcp add segmentstream \ --transport http \ https://mcp.segmentstream.com/mcp
On first use, you’ll be prompted to authenticate via your browser.You can scope the server to your project or your user profile:
Flag
Scope
Config file
-s project
Shared with your team via .mcp.json
.mcp.json in project root
-s user
Personal, across all projects
~/.claude/settings.json
(default)
Local to this project
.mcp.json in project root
Open Settings → MCP and add a new server, or edit .cursor/mcp.json in your project root:
After setup, test that the connection works by asking your AI assistant:
List my SegmentStream projects
The assistant should call the list_active_projects tool and return a list of projects you have access to. If you see your projects, the connection is working.
