Questions

macOS 26 or later, on Apple silicon.

Not yet. macOS comes first and gets the full feature set; an iOS version is planned but not released, and nothing on this site should be read as promising a date for it.

No, and none is planned. It is a native Mac application, which is most of the point.

No. Every connection type is built into the app. There is no Java, no Electron, no bundled browser, no plugin marketplace and no server component to run anywhere.

In the Keychain, not in the app's database. The connection library syncs through your private iCloud database, and secrets are stored separately in the iCloud Keychain — which is what lets them reach your other Mac without ever entering the synced library.

It connects to the servers you tell it to. Two other things are on by default and each has its own switch in Settings, under Advanced: crash reports, which carry the stack trace and your Hydra and macOS version and nothing that identifies you; and usage analytics, which record which features are used along with your Hydra and macOS version, device model, screen size, language and timezone, tied to a random ID created for that installation. Our own server, not a third-party analytics provider, and it also logs the IP the request came from. The optional email screen in the welcome flow is collected only if you fill it in.

What is never sent either way: connection names, hostnames, file paths, queries, folder listings, file contents and credentials. Analytics record that a feature was used, never what you used it on. There are no advertising identifiers and nothing goes to advertisers or data brokers. The full detail is on the privacy page.

The folder scanner's judgement about whether a found secret is real runs on an on-device model, so the contents of your config files are not sent anywhere to be classified.

Yes, in place. A text file on any supported service opens as a tab and saves back to where it came from, with encoding and line endings preserved. It uses whatever write guarantee the service actually offers — a posix rename over SFTP, an If-Match precondition on S3, an atomic replace locally — and tells you when a service cannot do better. FTP, for instance, has to delete the file before writing it, and the app says so first.

It covers the working-copy loop: stage by file or by individual hunk (unstage by file), discard hunks or files, commit, switch branches, pull, push, and browse recent commits with per-commit diffs. It runs on go-git compiled into the app, so it does not shell out to /usr/bin/git and works under the sandbox.

It is not a replacement for a dedicated Git client on the harder operations — interactive rebase, conflict resolution and submodules are not there.

No, and it is not planned. SMB2 and SMB3 only.

Probably. S3 is implemented directly against the API rather than through a vendor SDK, with custom endpoints and path-style addressing, which is what most S3-compatible services need. The same applies to WebDAV and Azure Blob Storage, including Azurite and sovereign clouds.

$99, once. That includes twelve months of updates, and every version released during that year is yours permanently. When the year ends the app keeps working exactly as it is. This is not a subscription, and nothing stops if you never pay again. Renewing costs the same $99 and buys another twelve months of updates. You can skip it for as long as you like without losing anything you already have. There is a fourteen-day trial and no account is required.

Yes. A SQLPro Studio license takes 50% off, and a license for any other SQLPro app takes 25% off. Enter your key on the upgrade page and it gives you a one-time promo code for the store.

Connections stop. Your library, your saved passwords and export keep working, so nothing you put into the app is held hostage.

Same developer. Hankinsoft Development has built and supported SQLPro Studio since 2015. Hydra Dev Studio is a separate application that applies the same approach to everything else a developer connects to during a day.