# StarRocks MCP for AI Agents AI Agent Connect > StarRocks lets your AI agent manage high-performance OLAP clusters, run complex SQL queries, and monitor data ingestion in real-time. It turns your analytical database into a conversational interface for data engineers and SREs who need to audit schemas, check node health, and pull insights without manual terminal work. ## Overview - **Category:** brain-trust - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_bhPoKhzny8myTwStxuxPah44hAT7Sn8O3RSNdBbS/ai-agent-connect - **Tags:** olap, distributed-database, data-querying, schema-management, cluster-monitoring, real-time-analytics ## Description StarRocks lets your AI agent manage your high-performance OLAP infrastructure. Managing a distributed analytical database usually feels like a full-time job of babysitting. You're constantly flipping between the StarRocks Manager, various terminal windows for SQL queries, and monitoring dashboards to see if your ingestion jobs actually finished. It's a lot of context-switching that pulls you away from actually solving data problems. This Connector changes that by letting your AI client do the heavy lifting for you. Instead of manual checks, you can just ask your agent to find specific tables, check the status of a backend node, or run a complex query to see how a specific metric is performing. It makes your data infrastructure feel like a part of your natural conversation. You can audit your entire cluster or grab a quick storage report without ever leaving your workspace. By connecting StarRocks to your workflow through the Vinkius catalog, you get a reliable way to keep your analytical platform running smoothly while staying focused on the high-level insights that matter most. ## Tools ### get_cluster_info Get frontend nodes info. This helps you check the status of your frontend infrastructure. ### list_nodes List backend nodes. Use this to check the health and status of your backend cluster. ### execute_query Execute arbitrary SQL query. Use this to pull real-time data insights directly into your chat. ### get_storage_usage Get data storage statistics. Use this to monitor disk usage across the entire cluster. ### get_table_schema Get table structure. This allows you to see the exact columns and types for any table. ### list_databases List all databases. Use this to see a high-level overview of your data environment. ### list_jobs List data load jobs. This helps you monitor ingestion progress and find failed tasks. ### list_mvs List materialized views. Use this to audit your pre-computed analytical views. ### list_tables List tables in a database. This lets you see every table within a specific database. ### list_views List database views. This helps you manage and audit your database views. ## Prompt Examples **Prompt:** ``` Check if our latest data ingestion jobs finished successfully. ``` **Response:** ``` I've checked your recent data load jobs. Here is the status: | Job Name | Status | Duration | | :--- | :--- | :--- | | `daily_sales_load` | ✅ Success | 4m 12s | | `user_logs_sync` | ❌ Failed | 1m 05s | The `user_logs_sync` job failed. Would you like me to check the logs for that specific job? ``` **Prompt:** ``` Show me the storage usage for the cluster. ``` **Response:** ``` I've retrieved the storage statistics for your StarRocks cluster: * **Total Used Space:** 4.2 TB * **FE Nodes:** 128 GB used / 512 GB total * **BE Nodes:** 3.8 TB used / 10 TB total Storage is currently at 38% capacity. Everything looks healthy. ``` **Prompt:** ``` List all databases and tell me which one has the most tables. ``` **Response:** ``` I've listed your databases: * `sales_analytics` (42 tables) * `user_behavior` (12 tables) * `logs_archive` (150 tables) * `marketing_data` (8 tables) The `logs_archive` database has the most tables with 150 total. ``` ## Capabilities ### Run arbitrary SQL queries Get real-time data results by asking your agent to execute complex queries. ### Monitor cluster health Check the status of frontend and backend nodes to ensure your infrastructure is up. ### Audit database schemas List all databases and retrieve table structures to keep your data mapped. ### Track ingestion jobs View the status and performance of your data load jobs in real-time. ### Get storage insights Retrieve disk usage and data size statistics across your distributed cluster. ## Use Cases ### Checking failed ingestion A data engineer needs to see if a specific job failed. They ask the agent to list jobs and identify the error. ### Troubleshooting cluster slowdowns An SRE notices a slowdown and asks the agent to check backend node health to find the bottleneck. ### Quick metric extraction An analyst needs a summary of sales. They ask the agent to execute a query on the sales table. ### Schema auditing A lead needs to audit the schema of a new database. They ask the agent to list databases and show the structure. ## Benefits - Stop manual terminal work by pulling insights directly into your chat without writing every line of SQL. - Keep your cluster healthy by checking node status and frontend info instantly through your agent. - Audit your data structures faster by listing databases and schemas in a single conversation. - Monitor your pipelines without refreshing pages by tracking load jobs and ingestion performance. - Prevent storage issues before they happen by checking disk stats across the whole cluster. - Manage complex views easily by listing all available materialized views and database views. ## How It Works The bottom line is you get a conversational interface for your entire StarRocks infrastructure. 1. Subscribe to the StarRocks MCP on Vinkius. 2. Enter your FE Host, Port, User, and Password in your AI client settings. 3. Ask your agent to query data, check node health, or list schemas. ## Frequently Asked Questions **Can I use StarRocks MCP to run SQL queries?** Yes, you can run arbitrary SQL queries directly through your AI agent. This allows you to pull real-time data insights without leaving your workspace. **Does StarRocks MCP help with cluster health?** Yes, it lets you check the status of both frontend and backend nodes. You can quickly identify which parts of your infrastructure are online or healthy. **Can I monitor my data ingestion with StarRocks MCP?** Yes, you can view your load jobs and their performance metrics. This helps you stay on top of your pipelines and identify failed tasks immediately. **Is StarRocks MCP good for auditing schemas?** Yes, it can list all databases, tables, and schemas instantly. It makes it much easier to audit your data structures and understand your environment. **How do I check storage space using StarRocks MCP?** You can ask your agent to retrieve disk usage statistics for the whole cluster. It provides a clear picture of your storage capacity across all nodes. **Can StarRocks MCP see my materialized views?** Yes, it can list all your materialized views and database views. This is helpful for managing your pre-computed analytics and ensuring they are up to date. **How do I find my StarRocks FE Host and Port?** The FE Host is the IP address or domain of your StarRocks Frontend node. The default HTTP port is 8030. You can verify these in your deployment configuration or by contacting your administrator. **Can I execute DELETE or UPDATE queries through this server?** Yes. Use the `execute_query` tool. As long as your authenticated user has the necessary permissions in StarRocks, you can execute any valid SQL statement, including DDL and DML operations. **Is it possible to monitor cluster health through the agent?** Yes! Use the `get_cluster_info` and `list_nodes` tools to retrieve real-time status for all Frontend and Backend nodes, helping you identify offline or overloaded nodes instantly.