DBQuill

DBQuill is an open-source desktop agent for querying databases in natural language, reviewing changes and viewing charts. It is best suited to Windows users working with SQLite or reading from supported MySQL and PostgreSQL databases. The guarded write workflow is its strongest reason to choose it; the source-first setup and narrow platform target are meaningful trade-offs.

Overview

DBQuill combines natural-language queries with database operations and visual results. It supports SQLite, with read paths for MySQL 8.4 and PostgreSQL 17. That distinction matters: users should not expect to make changes to remote MySQL or PostgreSQL databases. The app accepts OpenAI-compatible text-model endpoints, so it does not require a particular model vendor.

Its local-first approach keeps database credentials, model profiles, sessions, audit records and uploads on the user's machine, outside the source tree. Audit records retain controlled metadata and hashes rather than raw prompts, SQL, credentials or result rows. That is a considered boundary for sensitive work, though local storage alone does not make the app suitable for every security or deployment requirement.

Key features

Queries, files and charts

Users can ask questions in natural language and view results as charts. CSV and .xlsx attachments are converted into a local SQLite database, bringing tabular files into the same workflow without requiring a separate database server. Legacy .xls is unsupported, so older spreadsheet archives may need conversion first.

Controlled writes and bounded reads

Writes require validation, a change preview and explicit confirmation. Remote DDL is blocked, limiting one class of potentially destructive operation. These checkpoints suit users who want to inspect proposed changes rather than let an agent apply them silently; they also add friction for workflows that depend on unattended updates.

Reads are single-statement and row-bounded, while SQLite reads use physical read-only connections plus query_only. Those safeguards constrain the query path, but do not remove the need to review database access and permissions in the user's environment.

Pricing

DBQuill is free: its DBQuill plan costs nothing and is open source. The plan includes natural-language queries, write operations, a semantic layer and result visualizations, with self-hosted deployment. There is no paid tier or free-trial decision to weigh; the trade-off is that users take on a source-first installation rather than receiving a signed native installer.

Platforms

The current release targets Windows 10 or 11 x64 with CPython 3.12 and Microsoft Edge WebView2 Runtime. It is not yet a release target for macOS, Linux, Windows on ARM or other Python versions. Because distribution is source-first and there is no signed native installer, setup is better suited to users comfortable following installation guidance than to anyone expecting a conventional packaged desktop app.

Who it's for

DBQuill fits people who work with SQLite, need read access to MySQL 8.4 or PostgreSQL 17, or want to query tabular files while keeping data and audit records local. Its explicit review-before-write process is especially relevant when users want control over database changes. It is a poor fit for users who need remote schema changes, a broad operating-system choice, or a committed support response time.

Pros and cons

• Pros: The validation, preview and confirmation steps make writes deliberate rather than automatic.
• Pros: Local storage and audit records that omit raw prompts, SQL and result rows limit what the audit trail retains.
• Pros: CSV and .xlsx imports become local SQLite data, extending the tool beyond existing database connections.
• Cons: MySQL and PostgreSQL have read paths, while SQLite is the supported database for direct write use.
• Cons: Windows 10 or 11 x64 is the sole current release target, and installation is source-first without a signed native installer.
• Cons: Legacy .xls files are unsupported, and support offers guidance through installation materials, diagnostics and GitHub issues without a response-time or security-report bounty guarantee.

Alternatives

For a wider choice among products in this category, browse AI Database Assistants.

• DataZen is a free, GPLv3 option for Linux, macOS and Windows users who want a cross-platform alternative and can bring their own AI provider.
• Insight O' Mate may suit a founder working with one database who can stay within 20 analyses per day on its free plan, or who wants a paid plan with unlimited queries and advanced export.
• Outerbase AI is worth considering for a small group needing a web, API, self-hosted or desktop option; its free plan caps usage at 5 users, 1 base and 10 EZQL queries per month.
• YourQL is another free desktop application for Linux, macOS and Windows, though it is described as a work in progress.
• Florentine.ai is a web and API alternative for MongoDB or MySQL users; its free plan allows 1 database, 5 active collections or tables and 500 requests per month.
• Wren AI offers an open-source context engine for individual developers through CLI and MCP without a UI, making it a different fit from a desktop app.
• DataLine is a free open-source project for users seeking an alternative with Linux, macOS, Windows, web or self-hosted availability.
• Kyligence is a paid alternative with a free trial for users whose platform requirements match its API, Linux, self-hosted or web offerings.

Verdict

Choose DBQuill if you use Windows 10 or 11 x64, want natural-language work across SQLite and read-only MySQL or PostgreSQL paths, and value explicit approval before changes. Its main advantage is the combination of local handling and guarded writes; look elsewhere if you need a signed installer, macOS or Linux support, remote DDL, or a defined support commitment.