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

# Overview

> Learn how to connect AI agents to Profound

Connect AI tools and agents to Profound using the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction) open source standard for connecting AI applications to external systems.

## What is Profound MCP?

Profound MCP is our hosted server for integrating with AI tools that support MCP. Instead of wiring up a custom integration from scratch, you connect an MCP client to Profound and let the client call tools to execute API operations with your Profound account.

### Why use Profound MCP?

Profound MCP lets you connect to Profound quickly, without building a custom integration. Your Profound data and Agents become available directly inside the AI tool you already use. See how to connect common MCP clients to Profound in [Connection guides](/mcp/common-mcp-clients).

### What can you do with Profound MCP?

* **Docs** — Search Profound docs, endpoint, and authentication patterns before you start building
* **Visibility Reports** — Retrieve visibility metrics and performance data for companies within specific categories
* **Sentiment Analysis** — Analyze sentiment data and emotional responses across companies and topics
* **Citation Reports** — Track mentions, references, and citation patterns across domains and pages
* **Raw Data Access** — Access unprocessed data for custom analysis and reporting
* **Agent Analytics** — Access bot reports, referral reports, and raw analytics data
* **Build and Run Profound Agents** — Start existing Agents, create and publish new workflows, and poll runs for status and outputs

For the full tool and resource catalog, see [Analytics capabilities](/mcp/capabilities/analytics-capabilities) and [Agents capabilities](/mcp/capabilities/agents-capabilities).