Understanding the MCP Server

The Qualta MCP Server lets AI clients that speak the Model Context Protocol (MCP) — e.g., Claude Cowork, ChatGPT Work, custom agents — securely read a team's Qualta Guides and their interviews as context. It is read-only: it retreives and searches data; it never creates or changes anything.

Data Access & Privacy

When you connect an MCP server to an AI agent, you’re granting that agent access to the data exposed by that server. Once accessed, this data is no longer under Qualta’s control — it becomes subject to the connected AI agent’s terms of service and privacy policy, not Qualta’s.

This means the agent (and its provider) determines how your data is stored, used, and protected after access. Qualta cannot enforce its own privacy or security standards once the data leaves our system through an MCP connection.

Only connect MCP servers to AI clients you trust. Before granting access, review the AI provider’s terms of service and privacy policy to understand how they’ll handle your data. Treat each connection as a deliberate choice to share data with a third party.

How to Connect

Clients connect to the MCP server over streamable HTTP. Point your client’s MCP configuration at the following endpoint:

https://qualta.io/mcp/qualta

Authentication

Authentication is required — an unauthenticated request is rejected (HTTP 401) with a challenge that tells compliant clients how to authenticate.

OAuth 2.1 (Recommended)

Your client discovers the server’s OAuth settings automatically. You will be redirected to a Qualta authorization screen to approve or deny access. On approval, the client receives an access token allowing it to call MCP tools and resources.

More information on connecting to remote MCP servers can be found here.

API Token

For programmatic access, generate an API token with the mcp:use  permission (see How to Create an API Token), then send it as a bearer token in the Authorization  header with each request.

When authenticating via an API token, ensure the following request headers are set: 

• Accept: application/json
• Content-Type: application/json
• Authorization: Bearer <token>

For more information on programmatic connections to an MCP server, refer to the Build an MCP client tutorial and Specification documents on the official MCP documentation site.

What a Connected Client Can Access (Scoping & Permissions)

The Qualta MCP server does not provide access to any data beyond what you can already access in Qualta. Your existing Qualta permissions determine what data is available through the MCP server.

The MCP server can access:

• Teams: Only teams you belong to.
• Guides: Guides you are permitted to view, subject to your team role and project-level access.
• Interviews: Completed, non-hidden interviews.

MCP is available across Qualta plan tiers, but underlying Guide and Interview access still follows  normal Qualta rules, including an active subscription or trial, plus team and project permissions. If you don’t have access to the requested data, or if no data matches your request, the MCP server returns an empty result rather than an error.

Server Tools

Server tools are actions that a connected AI client can request the Qualta MCP server to perform. The available tools are listed below.

Server Resources

Server resources provide information that a connected AI client can access from the Qualta MCP server. The available resources are listed below.

Example Usage

Once the MCP server is installed in your AI client, you may start querying your interview data.

• "Summarize [team]'s Qualta interviews about [subject]"

Example: "Summarize Peach Computers Qualta interviews about EarBuds"

• "Use Qualta to determine what [demographic] think about [subject]"

Example: "Use Qualta to determine what women in their 30s think about EarBuds"

• "How does [external factor] relate to Qualta interviews on [subject]"

Example: "How does AI adoption relate to Qualta interviews on Peach computers"

• “What are the most common themes in Qualta interviews about [subject]?”

Example: “What are the most common themes in Qualta interviews about EarBuds?”

• “How do [demographic A] and [demographic B] differ in their views on [subject]?”

Example: “How do women in their 30s and men in their 30s differ in their views on EarBuds?”

• “What concerns do Qualta interview participants have about [subject]?”

Example: “What concerns do Qualta interview participants have about EarBuds?”

Note that quicker responses may be obtained by providing a Team ID and/or Guide ID up front. If you only want responses from a single Team across sessions, you may add the following instruction to your AI agent’s context (e.g., AGENTS.md , CLAUDE.md ) or a custom AI skill:

When using tools from the Qualta MCP server, always use this team ID { "teamId" : "YOUR_TEAM_ID" }

See How to Find Resource IDs to locate the appropriate Team ID.