# Redash MCP > Redash MCP lets your AI agent interact with your data warehouse using natural language. You can manage, run, and visualize SQL queries and dashboards without opening a browser tab. It connects raw database output directly to your chat conversation. ## Overview - **Category:** data-analytics - **Price:** Free - **Tags:** sql-queries, data-visualization, dashboards, redash-api, data-insights ## Description This connector brings the power of structured data analysis into your chat window. Instead of logging into Redash, navigating complex menus, or writing boilerplate code just to check a metric, you tell your agent what you need. The system handles running the necessary SQL queries, fetching fresh results, and pulling up dashboard visualizations—all conversationally. It gives you full control over your data environment. You can ask it to list all existing reports, run a new query using specific parameters, or even check the status of a long-running job that took hours. For visibility into what's available, the entire Vinkius catalog ensures you connect once and get access to this powerful data tool alongside dozens of others. It turns complex, multi-step reporting into simple back-and-forth conversation. ## Tools ### archive_dashboard Moves an existing dashboard into an archived state, removing it from active view. ### archive_query Archived a specified query object so it's no longer available for use. ### create_dashboard Builds and saves an entirely new dashboard container within the Redash environment. ### create_query Creates a brand-new, blank query object ready for you to write SQL against. ### execute_query Starts running a specific query, handling parameters if needed, or fetching stored results immediately. ### get_cached_query_result Retrieves the previously saved result for a non-parameterized query ID. ### get_dashboard Fetches all the specific details and structure of one particular dashboard. ### get_data_source Returns detailed information about a connected data source, like credentials or type. ### get_job Checks the current progress and status of any long-running query job. ### get_query_result Retrieves a complete result set using only its unique result ID number. ### get_query Fetches the full object details, including the SQL code, for an individual query. ### list_dashboards Returns a list of all dashboards you have access to in Redash. ### list_queries Provides a comprehensive list of every query saved within your account. ### test_data_source Runs a connection test to ensure the data source is online and accessible. ### update_dashboard Edits an existing dashboard object, allowing you to modify its structure or widgets. ### update_data_source Modifies the configuration settings of a data source, such as updating credentials. ### update_query Makes changes to an existing query object, including adjusting its SQL code or parameters. ## Prompt Examples **Prompt:** ``` List all my Redash queries. ``` **Response:** ``` I've retrieved your queries. You have 12 active queries, including 'Monthly Revenue' (ID: 45) and 'User Growth' (ID: 82). Would you like to see the details of a specific one? ``` **Prompt:** ``` Execute query ID 45 and show me the fresh results. ``` **Response:** ``` Starting execution for query 45... The job is finished. Here are the results: Total Revenue for October was $124,500 across 1,200 transactions. ``` **Prompt:** ``` Show me the 'Sales Overview' dashboard details. ``` **Response:** ``` I've fetched the 'Sales Overview' dashboard (slug: sales-overview). It contains 4 widgets: Revenue Chart, Regional Breakdown, Top Sales Reps, and Monthly Target Progress. ``` ## Capabilities ### Run real-time reports You can initiate a query run on demand and receive the resulting dataset in your chat. ### Manage saved queries List, create, update, or archive specific SQL code blocks to keep your workspace organized. ### View dashboard layouts Retrieve the structure and details of existing dashboards without manually building them. ### Monitor background jobs Check the status of complex or long-running data tasks to know when results are ready. ### Inspect data connections Verify the details and connection health of all underlying data sources. ## Use Cases ### The Analyst Needs a Quick Metric Check A data analyst needs to check last week's sales figures but doesn't want to rebuild the full dashboard. Instead of opening Redash, they simply ask their agent to execute a query on demand using `execute_query` and get the specific numbers right away. ### The BI Lead Needs Status Confirmation A business intelligence lead scheduled a massive report overnight. They don't know if it finished or failed. Rather than checking email digests, they ask their agent to check the job status using `get_job`, getting an instant confirmation of success or failure. ### The Engineer Needs Connection Validation A data engineer is setting up a new client connection and needs to validate the database link. They use their agent to run `test_data_source` first, confirming connectivity before writing any code. ### The Team Needs Report Cleanup After a project ends, a team member needs to declutter the workspace by removing old reports. They ask their agent to list all queries and then use `archive_query` on the ones no longer needed. ## Benefits - Eliminate context switching. You run queries and check dashboard status without ever leaving your conversational AI agent. - Manage the entire lifecycle of a report, from creating a new query object to running it and archiving it later, all through natural language commands. - Get immediate data visibility by using `execute_query` to fetch fresh results or `get_cached_query_result` for quick checks on non-parameterized data. - Maintain an organized workspace. Use tools like `list_queries`, `archive_query`, and `create_query` to keep your report library clean and functional. - Troubleshoot complex reports easily. You can use `get_job` to monitor long-running tasks, knowing exactly when the final results are ready for viewing. ## How It Works The bottom line is that it makes querying your business data feel as natural as sending an email. 1. Subscribe to this MCP and provide your Redash Base URL and API Key. 2. Direct your AI agent to perform a task, such as 'Show me the revenue for Q3.' 3. The system executes the necessary queries and presents you with the structured results directly in your chat. ## Frequently Asked Questions **How do I run a query using the Redash MCP?** You initiate a run by asking your agent to execute the report you need, often referencing its name or ID. The system then handles running the job and presenting the data. **Can I see all my dashboards with the Redash MCP?** Yes. You can ask your agent to list all available dashboards using `list_dashboards`, giving you a full overview of what reports exist for your team. **What if a query is running and I need to know the status? (Redash MCP)** Use the agent to check job status with `get_job`. This tells you immediately if the query finished, failed, or is still processing. **Does Redash MCP help me keep old queries organized?** Absolutely. You can use `archive_query` to remove outdated code from your active list and `list_queries` to see what you've kept. **Can the MCP handle parameterized reports?** Yes, the system handles parameters when running queries. You just need to provide the necessary variables in your request so it can fetch accurate, customized data.