# Tinybird Data Platform MCP for AI Agents AI Agent Connect > Tinybird Data Platform MCP. Connect your agent to Tinybird to manage real-time data pipelines, run SQL queries against your workspace, and audit your analytical infrastructure. It lets you see how data flows from ingestion to your final endpoints without leaving your chat interface. ## Overview - **Category:** loved-by-devs - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_e9MOySIFrTyItsoZeoVkFv0IVVN3MSGs8AJqsvq5/ai-agent-connect - **Tags:** real-time-data, data-ingestion, sql-queries, data-pipelines, api-endpoints, analytics-infrastructure ## Description This Connector lets you interact with your Tinybird infrastructure using natural language. Instead of jumping between tabs to check your data sources, inspect your transformation logic, or run manual queries, you can do it all through your agent. You can pull the exact SQL logic for a specific pipe to see how data is being processed, check ingestion stats to monitor performance, or run queries against your workspace to explore your data. It turns your agent into a hands-on operator for your real-time data flows. Since Vinkius hosts this in a unified catalog, you can manage these pipelines alongside other tools in your stack without switching environments. You get to see the how and the why of your data flows instantly, making it much easier to debug issues or verify outputs on the fly. ## Tools ### execute_sql_query Run an arbitrary SQL query against your Tinybird workspace to explore your data. This is useful for quick ad-hoc analysis. ### get_datasource_details Get comprehensive information for a specific Data Source. Use this to check configurations and metadata. ### get_pipe_details Get detailed information for a specific Pipe. This helps you understand the structure of your data flows. ### list_datasources Retrieve a list of all Data Sources in your current workspace. Use this to get an overview of your ingestion points. ### list_auth_tokens Retrieve a list of all authentication tokens in the workspace. Use this to audit your security and access. ### list_workspaces Retrieve a list of available workspaces. Use this to see all environments you have access to. ### query_pipe_data Execute a Pipe and retrieve the results as JSON. This is great for getting actual data outputs for your app. ### get_datasource_stats Retrieve ingestion and usage statistics for a Data Source. This helps you monitor performance and volume. ### list_pipe_nodes List all SQL nodes within a specific Pipe. This is the best way to see the exact logic of a transformation. ### list_pipes Retrieve a list of all Pipes in your current workspace. Use this to see all your active data flows at once. ## Prompt Examples **Prompt:** ``` Show me all the data sources I have in my workspace. ``` **Response:** ``` I'm looking into your workspace now. You have 3 active Data Sources: * **user_events** (4.2M rows) * **stripe_payments** (120k rows) * **ad_clicks** (8.9M rows) Which one do you want to see more details on? ``` **Prompt:** ``` What's the SQL logic for the 'daily_revenue' pipe? ``` **Response:** ``` Here is the transformation logic for the `daily_revenue` pipe. It uses 3 SQL nodes: 1. Filter raw payments by date 2. Group by currency 3. Sum the total amount Would you like me to run this pipe and show you the results? ``` **Prompt:** ``` Check the ingestion stats for the 'user_events' source. ``` **Response:** ``` I've pulled the latest stats for `user_events`. * **Ingestion Rate:** 450 rows per second * **Total Storage:** 12GB * **Row Count:** 4,200,500 Let me know if you need stats for any other sources. ``` ## Capabilities ### Run SQL queries against your workspace Execute arbitrary SQL queries to explore your data directly through your agent. ### List all data sources Get a full list of all data sources in your current workspace. ### View data source details Pull specific configuration information for any data source. ### Inspect pipe nodes See the exact SQL logic and nodes within a specific pipe. ### Check ingestion stats Retrieve usage statistics and row counts for your data sources. ### Execute pipes for JSON results Run a pipe and get the results back as JSON for easy data inspection. ### Audit auth tokens List all authentication tokens to manage your workspace security. ## Use Cases ### Debugging a broken pipe A data engineer notices a spike in errors and asks the agent to list the SQL nodes in a specific pipe to find the broken logic using list_pipe_nodes. ### Quick data validation A product analyst needs to know the total revenue for today and asks the agent to run an SQL query instead of building a new dashboard. ### Security auditing A DevOps lead asks the agent to list all auth tokens to ensure no old keys are still active in the workspace using list_auth_tokens. ### Infrastructure check A developer wants to see how many rows have been ingested into a specific source and asks for the ingestion stats using get_datasource_stats. ## Benefits - Skip the dashboard by running SQL queries directly through your agent using execute_sql_query. - Audit your entire data flow by listing and inspecting every Pipe and Data Source in your workspace. - Debug complex transformations faster by pulling specific SQL nodes with list_pipe_nodes. - Monitor your infrastructure health by checking ingestion stats and row counts with get_datasource_stats. - Secure your environment by listing and auditing all active authentication tokens using list_auth_tokens. - Get immediate JSON results from your data flows by running pipes directly with query_pipe_data. ## How It Works The bottom line is you get a chat-based command center for your real-time data infrastructure. 1. Subscribe to the Tinybird Data Platform MCP on Vinkius. 2. Provide your Tinybird Admin Token from your workspace settings. 3. Ask your agent to query data, audit pipes, or check ingestion stats. ## Frequently Asked Questions **Can I use the Tinybird Data Platform MCP to run SQL queries?** Yes, you can execute arbitrary SQL queries against your workspace directly through your agent. This lets you explore your data and run ad-hoc analysis without leaving your chat interface. **Does the Tinybird Data Platform MCP let me see my data pipelines?** It allows you to list all your Pipes and inspect the individual SQL nodes that make up each transformation. This is great for seeing exactly how your data is being processed. **How does the Tinybird Data Platform MCP help with data engineering?** It helps you audit your infrastructure by quickly listing sources, checking ingestion stats, and reviewing pipe logic. It streamlines the way you debug and monitor your real-time data flows. **Can I use the Tinybird Data Platform MCP to check my security tokens?** Yes, the Connector can list all authentication tokens in your workspace. This makes it easy to audit your access controls and ensure your environment is secure. **Does the Tinybird Data Platform MCP work with ClickHouse?** Yes, it supports running SQL queries using the ClickHouse dialect. This allows you to interact with your Tinybird data using the same syntax you already know. **Can I get JSON results from my Tinybird pipes?** You can execute a Pipe and receive the results back as JSON. This makes it easy to verify the final output of your data flows and use that data in your own applications. **Where do I find my Tinybird Admin Token?** Log in to your [**Tinybird Dashboard**](https://app.tinybird.co/), select a workspace, and go to the 'Auth Tokens' section. You can use the Admin Token or create a custom one with the necessary scopes. **Which SQL dialect does Tinybird use?** Tinybird is powered by ClickHouse, so it uses the ClickHouse SQL dialect, which includes powerful functions for high-performance analytical processing.