FAQ

Sensible questions,
honest answers.

Naggler is a small app with a few knobs worth explaining. Start with how Naggy's brain works — that's the one people ask most.

Naggy's brain

You have four options
to power Naggy.

Every option runs the same Naggy. They differ in where the thinking happens, who pays for it, and how private the conversation stays. Most people pick Cloud and never look back. Pick whichever fits.

Default for new users
Naggy emerging fresh from the cloud
Zero setup

Naggler Cloud

Ready the moment you install. Works out of the box. We cover the compute, served through OpenRouter with Zero Data Retention routing and a bounded model fallback chain.

  • Setup: None. It just works.
  • Cost to you: Free during beta.
  • Where thinking happens: Our Cloudflare relay routes every supported app version through OpenRouter and requires a Zero Data Retention endpoint.
  • Powered by: Models developed by U.S. companies—OpenAI, Anthropic, xAI, and Google—routed through OpenRouter with Zero Data Retention endpoint requirements. Model origin does not guarantee U.S. processing.
Naggy in developer mode, bring-your-own-key
Your key, your billing

Bring Your Own Key

Drop in your own API key and Naggy uses that instead. Usage is subject to your provider's billing, rate, and usage limits. Handy if you already have an Anthropic, OpenAI, or OpenRouter account, or want control over the spend.

  • Setup: Paste a key in Settings → AI Setup.
  • Cost to you: Whatever the provider charges — typically pennies per day of active use.
  • Where thinking happens: The provider's servers. We never see it.
  • Compatible with: Anthropic / OpenAI / OpenRouter keys today.
Naggy the tech wizard, configured with Ollama
You choose the server

Run Naggy with Ollama

Naggler sends each AI request to the Ollama server URL you configure. It stays on this device only when that server runs on this device — ideal when you want to run your own model locally.

  • Setup: Click "Download Ollama" in Settings to open Ollama's official download page, finish its installer, then choose a local model in Naggler.
  • Cost to you: Zero. You're using your own CPU/GPU.
  • Where thinking happens: At your configured Ollama server; it can run fully offline when that server and model are on this device.
  • Trade-off: Slightly slower and a bit less sharp than Cloud on older laptops.

The fourth option: a local coding agent. If the Claude Code CLI is already installed on your computer, Settings → AI Setup offers one more route — Naggy's replies come from that claude command, under whatever account it is signed in to. A local program is not local thinking: the request still goes to Anthropic, billed to that account rather than to us. It is not shown on computers without the CLI.

You can switch between any of these at any time in Settings → AI Setup. Your task ledger stays local and untouched either way.

Everything else

Other sensible questions.

Do I need an account?

No account, ever. During the beta you do hand over an email once — that's how Naggy sends you the install links and how we reach testers for important build or security updates — but there's nothing to log into, no password, no profile. Launch the app and you're running. Your task ledger is stored locally on your machine.

What does Naggler cost?

Naggler itself is free during beta. If you use Naggler Cloud, we cover the compute — you pay nothing. If you Bring Your Own Key, you pay your model provider directly (usually pennies per day). If you run Ollama locally, it's free — you're using your own hardware. If you point Naggy at the Claude Code CLI on your computer, you pay us nothing and the usage lands on whatever Claude account that CLI is signed in to.

What data leaves my machine?

Your task ledger lives on your disk. When you use an AI feature using Cloud or BYOK, Naggler sends the current request content (a typed message, or an app-generated heartbeat/title prompt) plus recent conversation, recent conversation, and the task/activity context needed to answer; see the exact context described on our Privacy page. With AI turned off, that context stays local. With the local coding agent, Naggler hands that copy to the claude CLI on your computer, which sends it to Anthropic under the account that CLI is signed in to. With Ollama, it goes to the Ollama server URL you configure and stays on this device only when that server runs on this device.

Which operating systems does it run on?

Windows 10+ and macOS. Naggler is desktop-only — there's no mobile app, and that's intentional. The tray icon on your desktop is the product.

Why does my computer warn me about running it?

Windows builds are signed with Azure Artifact Signing. Very new releases can still trip a SmartScreen reputation warning while install volume builds, but the unknown-publisher wall is no longer the expected path. macOS downloads are published only after Developer ID signing, notarization, and stapling pass; if you're testing a private pre-notarized build, macOS may still ask you to right-click → Open. Want to verify a checksum for your build? Reply to your download email — or write feedback@naggler.com — and we'll send the current values.

What should I expect on first install?

Windows — if the download looks stuck: Chrome and Edge hold new installers for confirmation, so you may see an Unconfirmed … .crdownload file instead of NagglerSetup.exe. Open your browser's downloads list and choose Keep. If Windows shows a blue SmartScreen box, click More info → Run anyway — the installer is signed, new releases just take time to build reputation.

Mac — install it, don't just open it: drag Naggler onto Applications, eject the DMG, then launch it from Applications or Spotlight rather than from inside the DMG.

Can I download it without giving you an email?

Not during the beta. The email form in the download section is the one way in — Naggy mails you the current signed builds for Windows and macOS, usually within a minute. Keeping the beta invite-by-email means we can reach testers for important build and security updates. It is not a marketing newsletter, and feedback@naggler.com takes you off the list any time.

Can I export my data?

Yes. Naggler's ledger is a local file you own. You can export check-ins, rankings, and session history to CSV or JSON from Settings → Export. No lock-in, nothing held hostage.

Does it work offline?

Yes, if your configured Ollama server and model run on this device. The core tracking loop — set priorities, answer pings, review your day — works without a network either way. You need a connection when you're talking to Naggy on Cloud, BYOK, the local coding agent, or a remote Ollama server.

Will you ever sell or train on my data?

No. We don't sell user data. We don't train models on your conversations. See our privacy policy for the detailed version and for information on how third-party processors handle your data.

Still stuck? Email feedback@naggler.com. A real person — occasionally Naggy himself — will get back to you.

Get Naggler