Schema-aware text to SQL

Plain English in.
Reviewable SQL out.

Grounded in your real schema and project rules, so what comes back is SQL you can read before running it.

New customers
validated
prompt

Show me all customers who signed up last month

generated SQL PostgreSQL
1SELECT *2FROM customers3WHERE created_at >= DATE_TRUNC(&#39;month&#39;, CURRENT_DATE - INTERVAL &#39;1 month&#39;)4  AND created_at < DATE_TRUNC(&#39;month&#39;, CURRENT_DATE);

Generates dialect-aware SQL for

PostgreSQL MySQL SQL Server Snowflake BigQuery Databricks PostgreSQL MySQL SQL Server Snowflake BigQuery Databricks

From a question to a query you can run.

TTSQL treats every prompt as a database workflow rather than an autocomplete: read the question, narrow it with schema context, apply the project's rules, then draft SQL that can be reviewed and corrected.

Connect your database

Securely link a database. TTSQL maps tables, joins, and schema context automatically.

tables, joins, column types

Ask in plain English

Type your question naturally. Our AI understands context, relationships, and business rules.

"revenue by plan this month"

Get SQL you can review

Receive validated SQL ready to review and run, with correction loops when static checks identify an issue.

SELECT ... GROUP BY ...

Platform

Built for real databases, not toy prompts.

Give TTSQL the structure around your data and every generated query has something to lean on.

Schema-aware generation

Use table names, relationships, sample queries, and project knowledge before drafting SQL.

AI-built dashboards

Describe what you want to see and get a whole dashboard generated from your schema: metrics, trends, breakdowns, filters, and drill-down.

describe "signups by week, revenue by plan,
and where churn is coming from"

Multi-engine SQL

Generate dialect-aware SQL for operational databases and warehouse-style analytics workflows.

Scoped data access

Keep database connections organized by project, with encrypted credentials and user-scoped workflows.

Correction loop

Use validation feedback to revise unsafe or invalid drafts before you review them.

Usage visibility

Track requests, errors, generated SQL, and project activity from the dashboard and API flow.

Knowledge controls

Add table descriptions, business terms, example queries, and custom rules to guide the model.

# project knowledge
term "active user" is activity in the last 28 days
term "revenue" is invoices.amount where status = 'paid'
rule always exclude internal test accounts

TTSQL, native on your Mac.

A genuinely native macOS app, written in Objective-C and AppKit rather than Electron. The same prompt-to-SQL workflow, with your connections and history kept on your own machine.

Native AppKit No web wrapper
Universal Apple Silicon and Intel
554 KB Tiny download
Download for macOS Linux, coming soon

Universal binary. macOS 11 Big Sur or later. Version 1.0.0.

The TTSQL macOS app showing a natural-language prompt and the SQL it generated

Start small. Scale when query access becomes critical.

Clear limits, the same SQL engine, and an upgrade path for teams that need governance at scale.

Starter

Perfect for exploring the platform

$ 0 free
  • 20 requests/day
  • 1 project
  • PostgreSQL support
  • Basic analytics
  • Community support
Start free

Pro

For growing teams and businesses

$ 9.99 /month
  • 300 requests/day
  • Unlimited projects
  • All databases
  • Advanced analytics
  • Priority support
  • Custom rules
  • API access
Upgrade

Enterprise

For large-scale deployments

Custom
  • Everything in Max
  • Unlimited requests
  • Dedicated support
  • SLA guarantees
  • Custom integrations
  • On-premise option
Contact sales

Answers before you connect a database.

How accurate is the AI-generated SQL?

Accuracy depends on schema quality and the context you provide. TTSQL improves results by grounding prompts in project knowledge, generated SQL review, and correction loops.

Can I connect multiple databases?

Yes! Pro and Enterprise plans support unlimited database connections across PostgreSQL, MySQL, MSSQL, and Oracle. Each project can have its own database connection.

Is my data secure?

Credentials are encrypted, projects scope database access, and generated SQL can be reviewed before execution. You control what databases and project knowledge are connected.

What happens if the AI makes a mistake?

Our self-correction engine can retry with validation context. If attempts fail, you get the error details for manual review. Corrections are not promoted into future examples automatically.

Can I customize the AI behavior?

Yes. Add custom business rules, table descriptions, and example queries to teach the model your domain-specific terminology on each grounded request.

Bring natural-language SQL into your product.

Self-serve via the dashboard, or reach out if you want dedicated support or a tailored deployment.