Expert SQL query generation for DBX Studio. Use when writing, optimizing, or debugging SQL queries against user database connections.
This project supports multiple database backends via user connections. Always write dialect-appropriate SQL.
| Dialect | Provider |
|---|---|
| PostgreSQL | Default / Railway |
| Snowflake | Via MCP connector |
| BigQuery | Via MCP connector |
| Databricks | Via MCP connector |
| MySQL | Via connection string |
| SQLite | Via connection string |
Always add LIMIT unless the user explicitly wants all rows:
SELECT * FROM "schema"."table" LIMIT 100;
WITH ranked AS (
SELECT *, ROW_NUMBER() OVER (PARTITION BY category ORDER BY created_at DESC) AS rn
FROM orders
)
SELECT * FROM ranked WHERE rn = 1;
SELECT
DATE_TRUNC('month', created_at) AS month,
COUNT(*) AS total,
SUM(amount) AS revenue
FROM orders
GROUP BY 1
ORDER BY 1 DESC;
SELECT
user_id,
amount,
SUM(amount) OVER (PARTITION BY user_id ORDER BY created_at) AS running_total
FROM transactions;
The AI has access to these tools — always use them rather than guessing:
| Tool | When to Use |
|---|---|
read_schema | First call — understand table structure |
get_table_data | Preview rows before writing complex queries |
execute_query | Run SELECT queries (SELECT/WITH only) |
describe_table | Get column details, FK relationships |
get_table_stats | Row counts, distributions |
generate_chart | Visualize query results |
execute_query"schema"."table"."column"read_schema or describe_table first