Umami MCP Server, developed by N A, is a free MCP server for the mcp platform. It bridges AI clients with Umami Analytics, facilitating natural language interaction with website data. This program comes in Python and TypeScript implementations, each offering distinct tools and distribution methods. It supports both stdio and HTTP transport protocols, making it compatible with various AI clients such as Claude Desktop, VS Code, and n8n. Users can monitor visitor activity, analyze user journeys, and create analytics dashboards directly through AI interactions, enhancing data accessibility for AI engineers and developers.

What does Umami MCP Server connect to?

The Umami MCP Server functions as a Model Context Protocol server, establishing connections with Umami Analytics instances. It provides AI clients with direct access to critical website analytics data. This server supports both self-hosted Umami Analytics versions and Umami Cloud, offering adaptability for various operational setups. Its core purpose is to enable AI clients to effectively process and interpret website performance metrics, user behavior, and visitor activity, transforming raw data into actionable insights for AI interactions.

Core features and tools

Umami MCP Server presents a diverse set of callable tools, varying across its implementations. The Python version offers 10 tools, including get_websites, get_website_stats, get_website_metrics, get_pageview_series, and get_active_visitors. It also includes get_session_ids, get_tracking_data, get_docs, get_screenshot, and get_html. A separate TypeScript implementation mentions 48 tools, with key capabilities like get_websites, get_stats, get_pageviews, get_metrics, and get_active. The Dockerized TypeScript version provides 8 distinct tools.

Who is it for?

Umami MCP Server targets AI engineers, developers, and users of AI clients aiming to integrate and analyze Umami Analytics data through natural language interactions. It serves individuals utilizing AI clients such as Claude Desktop, VS Code, Cursor, Windsurf, Zed, Smithery, Claude Code, and n8n. The server empowers AI to query and interpret website performance, user behavior, and current visitor activity. Skip this if your primary need is not to connect an AI client to Umami Analytics data for natural language processing.

How to get started

The server authenticates using an API key for Umami Cloud or a username and password for self-hosted Umami 3.x. For HTTP transport, credentials are passed via headers such as X-Umami-Api-Key, X-Umami-Username, or X-Umami-Password, while environment variables like UMAMI_URL can also be configured.

Use cases

• Asking an AI client to build a visual dashboard of website analytics.
• Analyzing user journeys to identify pain points and drop-off areas.
• Monitoring current visitor activity and comparing it with historical trends.

Umami MCP Server provides a vital bridge between AI clients and Umami Analytics, facilitating natural language data analysis. Its dual implementation approach offers flexibility in deployment and tool access. The server supports various authentication methods and transport protocols, making it adaptable for diverse AI engineering workflows. It empowers users to gain insights into website performance and user behavior directly through AI interactions.