> ## Documentation Index
> Fetch the complete documentation index at: https://docs.segmentstream.com/llms.txt
> Use this file to discover all available pages before exploring further.

# SegmentStream MCP

> Connect AI tools to your SegmentStream workspace using the Model Context Protocol (MCP) to query attribution data, manage configurations, and analyze marketing performance.

SegmentStream MCP is a hosted server that gives AI tools secure, read-only access to your SegmentStream workspace through the [Model Context Protocol](https://modelcontextprotocol.io/) — an open standard for connecting AI assistants to external data sources.

Once connected, your AI assistant can query attribution reports, inspect project configurations, explore user journeys, run BigQuery queries against your dataset, and more — all through natural language.

## Why use SegmentStream MCP?

<CardGroup cols={3}>
  <Card title="Natural language queries" icon="message-lines">
    Ask questions about campaign performance, attribution, and conversions in plain English instead of navigating dashboards.
  </Card>

  <Card title="Secure access" icon="lock">
    Authenticate with OAuth — the server enforces the same permissions as the SegmentStream UI.
  </Card>

  <Card title="Works with any MCP client" icon="plug">
    Compatible with Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, and any tool that supports MCP.
  </Card>
</CardGroup>

## What you can do

With SegmentStream MCP, your AI assistant can:

* **Query attribution reports** — pull campaign performance data by any dimension, date range, and attribution model
* **Analyze conversions** — check conversion statistics, trends, geographic breakdowns, and individual conversion records
* **Explore user journeys** — trace individual user paths and see how attribution credits are distributed
* **Inspect project configuration** — review data sources, attribution models, conversions, audiences, and grouped dimensions
* **Configure measurement** — create projects, connect data sources, define conversions and attribution models, edit channel groupings, and save reports
* **Run BigQuery SQL** — execute queries directly against your project's BigQuery dataset
* **Monitor data quality** — check cost data quality scores, workflow statuses, and incidents
* **Work with audiences** — list audiences, check memberships, and view inclusion statistics
* **Explore portfolio optimization** — review portfolio performance history, optimization scenarios, and budget allocation recommendations

<Note>
  Mutating tools (creating projects, conversions, attribution models, grouped dimensions, saved reports, and connecting data sources) honor the same permissions as the SegmentStream UI. Project members with viewer-only access can call read tools but not write tools.
</Note>

## Example prompts

Here are some things you can ask your AI assistant after connecting SegmentStream MCP:

* "Show me the top 10 campaigns by ROAS for the last 30 days"
* "What is the cost data quality score for our Facebook data source?"
* "List all conversions configured in the project"
* "Show me the user journey for anonymous ID `abc123` over the last 7 days"
* "Run a BigQuery query to count sessions by country for last week"
* "Which attribution models are configured, and which one is the default?"

## Get started

<CardGroup cols={3}>
  <Card title="Built-in measurement skills" icon="brain" href="/mcp/supported-tools">
    All measurement skills and commands are built into the server — nothing extra to install.
  </Card>

  <Card title="Getting started" icon="rocket" href="/mcp/getting-started">
    Connect any MCP-compatible tool in under a minute with OAuth.
  </Card>

  <Card title="Supported tools" icon="wrench" href="/mcp/supported-tools">
    Browse all available tools grouped by category with example prompts.
  </Card>
</CardGroup>