improvado-customer

Documentation

Documentation link · transport http

Tools

• clickhouseTool — Execute SQL against ClickHouse warehouse — customer workspace scope
• queryDestinationTool — Query downstream destinations directly (BigQuery, Snowflake, Redshift)
• visualizationTool — Generate charts — line, bar, area, pie, scatter, table
• heatmapVisualizationTool — Create heatmaps for time-based metric analysis
• sendEmailTool — Send reports and visualizations via email
• discoveryListDataSourcesTool — List all data sources available in the workspace
• discoveryListConnectorsTool — List connectors/integrations by data source name
• discoveryListAccountsTool — List ad accounts within a specific connector
• discoveryRequestTool — Execute live query against any of 1,000+ platforms (Google Ads GAQL, Meta, etc.)
• discoveryWebSearchTool — Search documentation and support resources for a data source
• getConnectionsTool — Get all connections for a datasource in current workspace
• getConnectionAccountsTool — Get ad accounts associated with a specific connection
• getDestinationConnectionsTool — List destination connections (where data lands)
• createExtractTool — Create a new scheduled data extraction from a source
• updateExtractTool — Update extract schedule, accounts, or field selection
• getExtractByIdTool — Get details of a specific extract by ID
• getExtractTemplatesTool — List available extract templates
• getExtractTemplateByIdTool — Get a specific extract template by ID
• listAllExtractsTool — List all extracts in the workspace with status
• pauseExtractTool — Pause a running extract (stops scheduled pulls)
• unpauseExtractTool — Resume a paused extract
• discoverFileFieldsTool — Discover available fields for a file-based data source
• createLoadTool — Create a new load to push data to a destination
• updateLoadTool — Update load destination or schedule settings
• getLoadTool — Get details of a specific load by ID
• getLoadsTool — List loads with filtering options
• listLoadsTool — List all loads in the workspace
• deleteLoadTool — Delete a load permanently
• pauseLoadTool — Pause a running load
• unpauseLoadTool — Resume a paused load
• runLoadTool — Manually trigger a load run immediately
• getPipelineRunsTool — Get run history for a pipeline
• getPipelineRunResultTool — Get detailed results of a specific pipeline run
• generateFlatDataSampleTool — Generate a data sample from a flat file source
• createCustomPipelineTool — Create a custom SQL/Python transformation pipeline
• updateCustomPipelineTool — Update custom pipeline logic or schedule
• deleteCustomPipelineTool — Delete a custom pipeline
• listCustomPipelinesTool — List all custom pipelines in workspace
• executeCustomPipelineTool — Execute a custom pipeline immediately
• runCustomPipelineCodeTool — Run pipeline code with custom parameters
• createMdgViewTool — Create a governance rule/view (naming, budget, classification)
• getMdgRuleLibraryExampleTool — Get examples from the 250+ pre-built MDG rule library
• testMdgNotificationTool — Test MDG alert and notification configuration
• previewEmailImportAddressTool — Preview email import address for data ingestion
• uploadFlatDataFileTool — Upload CSV or Excel file for processing
• uploadNewFileToExtractTool — Upload a new file version to an existing extract
• createFlatDataExtractTool — Create an extract pipeline from flat file data
• createRecipeTool — Create a new composable AI analysis notebook
• updateRecipeTool — Update recipe settings, cells, or parameters
• getRecipeTool — Retrieve a recipe by ID
• getRecipeTemplatesTool — List available recipe templates (attribution, reporting, etc.)
• createRecipeStepTool — Add a SQL or Python step to an existing recipe
• getRecipeStepTool — Get a specific step within a recipe
• listRecipesTool — List all recipes in the current workspace
• activateRecipeTool — Activate and run a recipe
• nameRecipeTool — Set or update the display name of a recipe
• createLesIntegrationChatTool — Create a LES integration chat session for guided setup
• scheduleChatTool — Schedule a recurring AI chat/analysis session
• getScheduledChatTool — Get details of a scheduled chat session
• deleteScheduledChatTool — Remove a scheduled chat session
• getDashboardTool — Retrieve a dashboard by ID
• listDashboardsTool — List all dashboards in the workspace
• getCustomDashboardsAndDecksPromptTool — Get context prompt for custom dashboards and decks
• createDocument — Create a new document in the workspace
• editDocument — Edit content of an existing document
• getDocument — Retrieve a document by ID
• deleteDocument — Delete a document permanently
• listDocuments — List all documents in the workspace
• findChat — Search past AI conversations by query
• getChat — Get full conversation context and history by ID
• getManagedConnectionCredentialsTool — Get connection credential metadata (IDs/status — never raw secrets)
• documentationTool — Query Improvado product documentation
• retrieveDocumentationTool — Retrieve specific documentation files by path
• listDataTablesTool — List all data tables available in current workspace
• getCurrentWorkspaceTool — Get current workspace context — agency, workspace ID, name
• switchWorkspaceTool — Switch to a different workspace within allowed agencies
• getWorkspacesTool — List all workspaces available to the authenticated user
• mcpListToolsTool — Dynamically list all available MCP tools at runtime
• mcpCallToolTool — Call any MCP tool by name (meta-wrapper for agents)
• mcpListResourcesTool — List MCP resources available in current context
• mcpReadResourceTool — Read a specific MCP resource by URI
• createConnectionTool — Create a new data source connection in the workspace
• customConnectorFormTool — Access custom connector configuration form

About MCP

The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.

Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.