Connect a database
Enter your PostgreSQL or MySQL credentials, or discover instances on an RDS host. The connection is tested and the schema read before anything else happens.
Database Agent turns your question into SQL, runs it read-only, and answers with the numbers, a chart and the exact query it wrote. Connect PostgreSQL or MySQL, describe your data once, and start asking.
Try Database Agentis pinned to the top of the page — it stays there as you scroll.
customers ordersSELECT c.first_name, c.last_name, SUM(o.total_amount) AS total_spent FROM customers c JOIN orders o ON c.id = o.customer_id GROUP BY c.id, c.first_name, c.last_name ORDER BY total_spent DESC LIMIT 5;
| first_name | last_name | total_spent |
|---|---|---|
| Priya | Raman | 11,984.20 |
| Arjun | Mehta | 9,210.75 |
| Sneha | Kulkarni | 7,488.90 |
| Vikram | Desai | 6,602.15 |
| Neha | Sharma | 5,922.60 |
How it works
There is no setup phase to get through first. Everything between a question and its answer — reading the schema, writing the SQL, running it, drawing the chart — happens on its own.
Three steps, none of which is a configuration screen.
Enter your PostgreSQL or MySQL credentials, or discover instances on an RDS host. The connection is tested and the schema read before anything else happens.
Database Agent drafts a description of the database from your schema. Keep it as-is, or reword it later if your team calls things something else — either way it is ready to answer.
Type a question. Get a heading, a plain-English answer, a result table and a chart — plus the SQL, row count and timing whenever you want to check the work.
Features
Everything you need to ask a question, check the answer, share it and watch how it performs over time — and nothing you don’t.
Every answer carries the executed SQL, the raw response, the tables it touched, timing and row count — and a one-click CSV export. Nothing is a black box you have to trust.
Add as many PostgreSQL and MySQL databases as you need, or discover several at once on an RDS host. Each keeps its own description, visible tables and connection status.
Choose which tables are visible, describe them in your own words, and switch a whole database off without deleting it. A table that is not visible cannot appear in a query.
When a result is easier to see than to read, Database Agent picks the chart and draws it beside the table — no chart builder, no configuration. The rows stay one CSV export away.
Schedule a set of questions to run together and read them as one grouped report — the same weekly numbers, answered without anyone opening the dashboard.
Query volume, success rate, response time and cost over any period, with the full query log behind every number.
Use cases
Every one of these is a question someone asks an analyst today, and waits a day for. The wording below is what you would type.
Answer the one-off before it becomes a ticket, and keep the SQL for the dashboard you build later.
Which acquisition channels drove the top 10% of revenue last quarter?
Month-over-month repeat purchase rate for the last six months.
Read the operational number yourself, at the moment you need it, without an export in between.
Which orders shipped late in the last 30 days, by warehouse?
Open invoices past 45 days, with the account owner on each.
Put the agent behind a shared link or a widget, so the team asks the database instead of asking you.
Which accounts opened more than five tickets this month?
Median first-response time by plan, this week against last.
Integrations
Drop a chat bubble on your website, send a full-page link, or call the same agent from your own code. Invite teammates as editors or viewers, issue API keys, and restyle the widget with a live preview before it goes out.
Connects to
Reach it from
Generate a key under Share → API. A key can read answers, never change settings.
import requests r = requests.post( "https://app.futuresmart.ai/nl2sql/api/v1/agents/NL2SQL_379EACDE/chat", headers={"x-api-key": "<YOUR_API_KEY>"}, json={"query": "Show products with stock quantity less than 10", "chart_output": True, "follow_up": True, "debug_info": True}, timeout=120, ) d = r.json() print(d["response"]) print(d["debug_info"]["sql_query"])
Security & deployment
We follow industry-standard practices to protect your data — and if that is still not close enough, the whole product can run inside your own environment.
We can install Database Agent inside your own cloud or data centre. Your database, your questions and your results never leave your network.
Questions become read-only queries. Every answer shows the exact SQL it ran, and the full query log keeps a record you can audit later.
Invite teammates as editors or viewers, issue and revoke API keys, and keep whole tables out of reach — with a full activity trail behind every one.
Need it inside your own environment? That is a conversation, not a different product.
Connect PostgreSQL or MySQL, ask your first question, and read the answer, the chart and the SQL that produced them. Try Database Agent is at the top of the page.
Prefer a walkthrough first? Book a demo.