Connecting to the Validatar MCP Server

Validatar exposes a Model Context Protocol (MCP) server that allows AI assistants and development tools to interact directly with your Validatar instance. Through MCP, you can browse your catalog, create and execute tests, run scripts, and manage metadata — all from within an AI-powered environment like Claude Code, Cursor, or any MCP-compatible client.

Prerequisites

Before connecting to the Validatar MCP Server:

• Your Validatar instance must be running a version that includes MCP support.
• Your user account must have an active session with appropriate permissions for the tools you intend to use.
• You need the base URL of your Validatar instance (e.g., https://your-instance.cloud.validatar.com).
• Your AI client must support the Model Context Protocol (MCP) with HTTP transport.

How MCP Works with Validatar

The Model Context Protocol is an open standard that allows AI assistants to interact with external systems through a defined set of tools. Validatar implements an MCP server that:

• Runs over HTTP with stateless transport
• Authenticates requests using your existing Validatar session credentials
• Exposes a set of read and write tools for interacting with your data catalog, tests, projects, and metadata
• Tracks all write operations as AI Actions that can be reviewed and reverted

Connection Configuration

Endpoint

The MCP server is available at:

{your_validatar_instance_url}/core/mcp

The /core prefix is the application's base API path — it is required in all MCP endpoint URLs.

Authentication

The MCP server authenticates requests using a Validatar User Token of type MCP. This is not the same as a standard API key or session cookie — you must create a dedicated MCP token in Validatar.

To create an MCP token:

1. Log in to your Validatar instance.
2. Navigate to your User Profile settings.
3. Under User Tokens, create a new token and set the type to MCP.
4. Copy the generated token value.

The token is passed to the MCP server via the x-val-api-token header:

x-val-api-token: <your-token>

Client Configuration Examples

Claude Code (CLI)

The fastest way to add the Validatar MCP server to Claude Code is with the CLI:

claude mcp add --transport http <your_validatar_instance_name> <your_validatar_instance_url>/core/mcp --header "x-val-api-token: <your-token>"

For example:

claude mcp add --transport http validatar-prod https://mycompany.cloud.validatar.com/core/mcp --header "x-val-api-token: abc123def456"

This registers the server in your Claude Code settings and makes all Validatar MCP tools available in your sessions.

Claude Code (Manual Configuration)

Alternatively, add the server directly to your .claude/settings.json or project-level settings:

{
  "mcpServers": {
    "validatar": {
      "type": "http",
      "url": "https://your-instance.cloud.validatar.com/core/mcp",
      "headers": {
        "x-val-api-token": "YOUR_MCP_TOKEN"
      }
    }
  }
}

Cursor

In Cursor's MCP configuration file (.cursor/mcp.json):

{
  "mcpServers": {
    "validatar": {
      "type": "http",
      "url": "https://your-instance.cloud.validatar.com/core/mcp",
      "headers": {
        "x-val-api-token": "YOUR_MCP_TOKEN"
      }
    }
  }
}

Generic MCP Client

Any MCP-compatible client that supports HTTP transport can connect. The required configuration is:

Verifying Your Connection

Once configured, verify the connection by asking your AI assistant to run a simple read-only operation:

1. Ask the assistant to list your projects or describe a known data source.
2. If the connection is successful, the assistant will return results from your Validatar instance.
3. If authentication fails, you'll receive an error — double-check your API key and instance URL.

Tool Access Control

Validatar administrators can control which MCP tools are available to specific users through token-based access control. If a tool is unavailable to you, contact your Validatar administrator to verify your permissions.

Tools are divided into two categories:

• Read-only tools — browse catalog objects, list tests, search documentation
• Write tools — create tests, execute scripts, update metadata, manage folders and projects

See MCP Tools Reference for the complete list of available tools and their capabilities.

AI Action Tracking and Revert

Every write operation performed through MCP is logged as an AI Action. This provides:

• A full audit trail of changes made through AI assistants
• The ability to revert any AI-initiated change using the revert_action tool
• Visibility into what was changed, when, and by which user session

When a write tool executes successfully, the response includes a revert_action_key that can be used to undo the change.

Troubleshooting

Next Steps

• Review the full MCP Tools Reference to understand what operations are available.
• Try browsing your catalog by asking the assistant to describe a data source or list tables in a schema.
• Create your first test through natural language by describing what you want to validate.