Skip to content

Privacy

Claudette is designed with a simple privacy principle: your data stays on your machine.

Claudette has no user registration, no login, and no accounts. You download the app and start using it — there’s nothing to sign up for.

Claudette is an independent, community-built tool. It is not affiliated with, endorsed by, or sponsored by Anthropic, PBC. “Claude” and “Claude Code” are trademarks of Anthropic, PBC; their use here is descriptive — Claudette orchestrates the official Claude Code CLI — and does not imply any partnership.

Claudette does not authenticate to Anthropic on your behalf. It spawns the official claude CLI you have installed locally as a subprocess; the CLI authenticates itself with whatever credentials you have configured. Claudette never reads, copies, or forwards your Claude OAuth tokens.

Per the Claude Code legal and compliance page, “advertised usage limits for Pro and Max plans assume ordinary, individual usage of Claude Code and the Agent SDK.” We recommend keeping default parallelism low (1–3 simultaneous agents) and treating heavier use as something you explicitly opt into. Responsibility for staying within your plan’s terms is yours.

Claudette does not operate any remote servers. Nothing you do in Claudette is transmitted to any servers managed by Claudette or utensils. There is:

  • No telemetry
  • No analytics
  • No crash reporting
  • No usage tracking
  • No session recording

All AI communication flows directly between your machine and Anthropic. Claudette runs Claude Code as a local subprocess — your prompts and code go straight to Anthropic’s API, governed by Anthropic’s privacy policy and your own API agreement.

Your Machine ←→ Anthropic API
Claudette (local app)

Claudette never proxies, intercepts, or stores your API traffic on any third-party server.

Beyond the Claude API traffic that flows through your locally-installed claude CLI, Claudette itself reaches a small number of external endpoints. None of these handle your prompts, code, or Claude credentials.

DestinationWhat it’s forWhen
github.com/utensils/Claudette/releasesApp update checks (Tauri updater)On app start
api.github.com/repos/utensils/claudette/releasesNightly build discoveryIf nightly channel selected
peonping.github.io/registry/Soundpack registry indexIf you open the soundpack picker
github.com/<repo>/archive/refs/tags/<tag>.tar.gzSoundpack downloadsOnly when you install a pack
huggingface.co/distil-whisper/distil-large-v3/...Local speech-to-text modelOnly the first time you use voice input
fonts.googleapis.comUI font (rendered in webview)While the app window is open

There is no telemetry endpoint, no analytics endpoint, and no crash-reporting endpoint. The destinations above exist to make features work, not to track you.

Claudette stores workspace metadata and chat history in a local SQLite database on your machine:

  • macOS: ~/Library/Application Support/com.claudette.app/
  • Linux: ~/.local/share/com.claudette.app/

This database contains workspace names, chat messages, and settings — all stored locally. You can back it up, delete it, or move it at any time.

Claude Code plugins may require their own secrets — API keys, tokens, etc. Claudette stores these in the same secure-storage object Claude Code itself uses (the macOS Keychain entry Claude Code-credentials, or ~/.claude/.credentials.json on Linux), but only under its own pluginSecrets namespace. Your Claude OAuth tokens (claudeAiOauth.*) are never read or written by Claudette’s plugin code.

The optional remote workspaces feature connects to servers you set up and control. Claudette never routes traffic through our infrastructure. The connection is:

  • End-to-end encrypted with TLS
  • Authenticated with trust-on-first-use certificate pinning
  • Your servers only — you run claudette-server on your own machines

Claudette is fully open source under the MIT License. The entire codebase is available for audit at github.com/utensils/Claudette.