Skip to main content
Technical reference for the Credible Data MCP server, available at https://<org>.data.credibledata.com/mcp.

​
Core Tools

​
suggestAnalysis

Converts natural language analysis requests into executable Malloy query code. Parameters:
  • primaryRequest (required): Natural language description
  • projectName (required): Project to search within
  • packageName (optional): Specific package scope
  • sourceName (optional): Specific data source/table
  • modelPath (optional): Path to specific .malloy model file
Parameter Hierarchy: Parameters follow a hierarchy: project → package → modelPath → sourceName. You can only specify a right-side parameter if you provide the left-side parameter. Response: Returns a MalloyQueryResponseDTO containing:
  • matchingResults: Array of query suggestions with generated Malloy query code
  • entities: Relevant dimensions, measures, and views you can use
  • primaryTransactionTable: The main data source
  • nextSteps: Guidance on how to use the results
  • docs: Malloy syntax documentation and examples

​
executeQuery

Executes Malloy queries against data models and returns results. Parameters:
  • projectName (required): Project containing the model
  • packageName (required): Package containing the model
  • path (required): Path to the .malloy model file
  • query (optional): Custom Malloy query code
  • sourceName (optional): Source name (required with queryName)
  • queryName (optional): Name of predefined query/view
Execution Modes:
  • Custom Query: Use query parameter with generated code from suggestAnalysis
  • Predefined Query: Use queryName and sourceName for existing views

​
Discovery Tools

  • credible_projectList: Lists all available Credible projects
  • credible_packageList: Lists all packages within a specific project
  • credible_packageGet: Lists all models within a specific package
  • credible_modelGetText: Gets the raw text content of a specific model file

​
Authentication

The MCP server uses OAuth 2.0 with organization-based access controls. Configure your MCP client with your organization’s endpoint for secure access to semantic models.

​
Error Handling

The server returns standard MCP error responses for invalid requests, authentication failures, and query execution errors. Refer to the MCP specification for error code details.
⌘I