Overview
Streamline access to customer intelligence with the Dovetail MCP server. Designed for efficiency and seamless integration, the Dovetail MCP server enables AI models to securely access and search your Dovetail workspace. The Model Context Protocol (MCP) is an open standard that allows AI assistants, like Claude Desktop or Cursor - to connect with other tools and services. By connecting to an MCP server you can transform an AI assistant from a helpful generalist tool into a a powerful, informed teammate capable of giving more grounded answers and performing more complex tasks. Dovetail’s MCP server allows you to connect AI assistants to your Dovetail workspace, allowing these assistants to query your research data, summarize insights, and find evidence directly within your AI chat interface, without manually copying and pasting information.What can you do with the Dovetail MCP?
Normally, you need to manually provide your AI assistant with context by pasting transcripts or information into your chat. With the Dovetail MCP server, your AI assistant gets read access to your Dovetail data, docs and data points. It’s able to search across your workspace, dive into specific transcripts, look for projects and find published docs and highlights. When you connect Dovetail to an MCP-compatible AI assistant, you can use natural language to interact with your customer data.- Ask questions about your data
Query your findings by asking things like “What are the most common complaints about our checkout flow?” or “What did users say about the new dashboard design?” - Generate summaries
Quickly get a high-level overview of a specific project, folder, or set of insights without leaving your AI tool - Find evidence instantly
Ask the AI to find specific customer quotes (highlights) or existing insights that support a hypothesis you are working on. - Draft content with context
If you are using an AI to write a product brief or a blog post, the AI can reference your actual Dovetail data to ensure the content is grounded in real customer data.
How it works
The Dovetail MCP server acts as a secure translator between your AI assistant and your data in Dovetail. When you connect your AI assistant to Dovetail’s MCP server, your AI assistant gets access to a whole suite of tools it can use to answer any questions or requests you send it. Your AI assistant will be able to use these tools to gather context and provide you with richer, more grounded answers.Security and permissions
The AI only has access to the information you can access. It can’t see projects or data that you don’t have permission to access, and it can’t make changes to anything in your Dovetail workspace. You maintain full control. You can connect or disconnect your AI assistant from Dovetail whenever you like.Get started with Dovetail’s MCP server
There’s a few ways to get set up with an MCP server. It is a technical process, so whilst you don’t need to be a developer to use it, you may want to ask for assistance from a developer or a technical teammate to install it.You can get set up in one of several ways:
- Connecting with an API token
Most AI tools will require that you generate a “Personal API Key” from your Dovetail account settings. - Connecting with a first party integration
Some tools (like Figma Make) natively support Dovetail as an MCP connector. You just need to login to your Dovetail account. - Running a local server
Some tools (like Claude) may require you to run a small script on your computer to get connected.
If you are ready to set this up, or if you want to pass the instructions to a teammate who can help, please refer to our more in-depth technical documentation:
View developer docs →
View developer docs →