# Redash MCP for AI Agents AI Agent Connect > Redash MCP lets you manage SQL queries, run visualizations, and monitor dashboards directly through your AI agent. It bridges the gap between raw data and actionable insights by letting your agent fetch fresh results, manage query objects, and check data source statuses without you having to switch tabs or perform manual copy-pasting. ## Overview - **Category:** data-analytics - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_rRCXZFALibUsor8N9CjThd5ZlbvdaN0QeviiaDQu/ai-agent-connect - **Tags:** sql-queries, data-visualization, dashboards, redash-api, data-insights ## Description This Redash MCP connects your Redash instance to any AI agent so you can interact with your data using natural language. It bridges the gap between raw data and actionable insights by letting you query metrics without the friction of a dashboard UI. You can ask your agent to find specific reports, run SQL commands with parameters, or fetch fresh results from your data warehouse. Because Vinkius hosts this Connector, you can integrate it into your existing workflow to manage query objects, monitor long-running jobs, and check data source statuses. It handles the repetitive work of browsing dashboards or archiving old records, letting you focus on the actual analysis instead of clicking through menus. Instead of manually hunting for a specific metric or waiting for a dashboard to refresh, your agent handles the heavy lifting of data retrieval. You get a direct line to your data warehouse that feels like a conversation, making it much easier to stay on top of your KPIs without the usual technical overhead. ## Tools ### list_queries See a list of all queries in your Redash instance. This allows you to browse your entire query library quickly. ### test_data_source Verify that a data source connection is actually working. Use this to troubleshoot connection issues before running reports. ### archive_dashboard Archive a dashboard to clean up your view. This helps keep your workspace organized by hiding old reports. ### archive_query Archive an old query to keep your workspace tidy. Use this to move inactive SQL objects out of your primary view. ### create_dashboard Create a new dashboard object in Redash. This allows you to build new visualization layouts using your agent. ### create_query Create a new SQL query object. Use this to save new queries directly through your AI client. ### execute_query Run a query and get results, including handling specific parameters. This is how you fetch fresh data on demand. ### get_cached_query_result Grab a previously saved result for a non-parameterized query. This provides an instant way to see data without re-running code. ### get_dashboard Fetch the details of a specific dashboard. Use this to see the configuration and widgets of an existing report. ### get_data_source See the details and configuration of a specific data source. This helps you verify where your data is coming from. ### get_job Check the status of a running query task to see if it's finished. Use this to monitor progress on heavy data fetches. ### get_query_result Retrieve the result of a query using its specific result ID. This is useful for grabbing specific data points from a previous run. ### get_query Get the details of a specific query object. Use this to inspect the SQL and metadata of any query in your account. ### update_dashboard Modify the configuration of an existing dashboard. Use this to change layout settings or update dashboard metadata. ### update_data_source Change data source settings like SSH tunnel configurations. This allows you to manage your backend connections programmatically. ### update_query Edit the SQL or metadata of an existing query. Use this to make quick updates to your saved queries without the UI. ### list_dashboards Get a paginated list of all dashboards in your workspace. This helps you navigate your visual reports efficiently. ## Prompt Examples **Prompt:** ``` What's the status of my 'Monthly Sales' report? ``` **Response:** ``` I checked the 'Monthly Sales' report (Query ID: 102). It's currently running and is about 75% complete. I'll let you know as soon as the results are ready. ``` **Prompt:** ``` Show me the widgets on the Executive Summary dashboard. ``` **Response:** ``` The 'Executive Summary' dashboard contains the following 5 widgets: - **Total Revenue** (Big Number) - **Regional Sales** (Bar Chart) - **Active Users** (Line Graph) - **Churn Rate** (Gauge) - **Target Progress** (Table) ``` **Prompt:** ``` List all my queries from last week. ``` **Response:** ``` I've pulled your query list. Here are the ones you worked on recently: - *User Retention* (ID: 44) - *Ad Spend Analysis* (ID: 56) - *Inventory Count* (ID: 89) Would you like me to run any of these for you? ``` ## Capabilities ### Execute SQL queries Run queries on-demand and get back fresh or cached results. ### Manage query objects Create, update, and archive SQL queries to keep your workspace tidy. ### Access dashboards Browse and retrieve specific dashboard configurations and visualizations. ### Monitor query jobs Track the progress of long-running executions in real time. ### Inspect data sources Check connection statuses and configure data source settings. ### Fetch cached results Quickly grab non-parameterized query results without re-running the code. ## Use Cases ### I need the latest revenue numbers An analyst asks the agent to use `execute_query` on a specific ID to show the latest results in a table. ### Clean up my workspace A BI lead asks the agent to use `archive_query` on all reports that haven't been touched in a year. ### Check the dashboard status An engineer asks if the 'Sales Overview' dashboard is still active and uses `get_dashboard` to see its widgets. ### Verify the database connection A dev asks the agent to use `test_data_source` on the production database to report any errors. ## Benefits - Stop tab-switching by using `list_queries` and `list_dashboards` to find what you need without leaving your chat. - Get faster insights by using `get_cached_query_result` to pull data instantly without re-running heavy SQL. - Keep your workspace clean by using `archive_query` and `archive_dashboard` to hide old reports. - Stay updated on heavy tasks by using `get_job` to monitor the progress of long-running data fetches. - Fix connection issues faster by using `test_data_source` and `update_data_source` to manage your backend links. ## How It Works The bottom line is you get a natural language interface for your entire Redash data environment. 1. Subscribe to the Redash MCP on Vinkius and provide your Redash Base URL and API Key. 2. Connect the Connector to your preferred AI client like Claude or Cursor. 3. Ask your agent to run reports, fetch metrics, or manage your dashboards. ## Frequently Asked Questions **Can I use Redash MCP to run SQL queries?** Yes. You can ask your agent to execute specific queries and get the results back directly in your chat. It handles the execution and brings the data to you. **Does the Redash MCP support parameterized queries?** Yes, it does. Your agent can handle queries that require specific parameters, like date ranges or user IDs, to get the exact data you need. **Can I manage my dashboards with the Redash MCP?** You can. Your agent can list your dashboards, retrieve specific details about them, and even archive old ones to keep your workspace organized. **How do I check if my data sources are connected?** Just ask your agent to check the status of your data sources. It can inspect the connections and even help you update configurations like SSH tunnels. **Can I see the results of a query without running it again?** Yes, if the query isn't parameterized, your agent can fetch the last cached result for you instantly. **Is the Redash MCP good for data analysts?** It's a huge time saver for analysts who need to pull metrics quickly. Instead of manual navigation, you can just ask for the data and get it immediately. **How do I run a query and get the results?** Use the `execute_query` tool. If the query takes time, it returns a `job_id`. You can then use `get_job` to check the status, and once successful, use `get_query_result` with the provided ID to fetch the data. **Can I update the SQL of an existing query through the agent?** Yes! Use the `update_query` tool by providing the Query ID and the new SQL string in the `query` parameter. You can also update the name or data source ID. **How do I find a specific dashboard and its widgets?** First, use `list_dashboards` to find the slug of the dashboard you need. Then, use the `get_dashboard` tool with that slug to retrieve all its configuration details and widgets.