Skip to content
dont-panic
Menu

Get started

Get started

There is nothing to sign up for and no installer to download: no release has been published. You build the app from a clone of the private repository on your own Apple-Silicon Mac.

  1. Step 1

    Request access

    The source repository is private. Send a request through the contact form and you will hear back.

  2. Step 2

    Set up

    make setup checks your toolchain (make doctor prints OK, WARN or FAIL with the fix), installs the JavaScript dependencies and creates apps/desktop/.env from the example file.

  3. Step 3

    Add your keys

    In apps/desktop/.env, set DEEPGRAM_API_KEY for Deepgram and one of OPENAI_API_KEY, GROK_API_KEY or ANTHROPIC_API_KEY for answers. Apple on-device speech recognition needs no key.

  4. Step 4

    Run it and grant permissions

    make dev starts the app. On first launch macOS asks for Microphone, Screen Recording and Speech Recognition access; the app needs all three for capture and transcription.

Quickstart

# Use the clone URL you receive when access is granted.
git clone <clone-url> dont-panic
cd dont-panic
make setup    # checks prerequisites, installs deps, creates apps/desktop/.env
make dev      # Vite on :1426 plus tauri dev
make run      # or: build an unsigned .app and open it

Keys (apps/desktop/.env)

  • DEEPGRAM_API_KEY: Deepgram streaming speech-to-text
  • ANTHROPIC_API_KEY, OPENAI_API_KEY or GROK_API_KEY: the LLM that writes answers
  • A built .app reads ~/Library/Application Support/dev.dontpanic.app/.env instead; make install-env copies your file there once (mode 0600, never overwrites)

What you need

  • A Mac with Apple Silicon
  • Full Xcode (not only the Command Line Tools)
  • Rust via rustup (the repository pins the toolchain)
  • Node.js 24 and pnpm 10
  • API keys only for the cloud providers you want to use

macOS permissions

Microphone
Captures your voice (You) for transcription
Screen Recording
Captures system audio (Them) through ScreenCaptureKit, and a screenshot when you fire Screen AI
Speech Recognition
Apple on-device transcription, with no data sent to a server
Calendars (optional build feature)
Lists upcoming events so meeting notes can be matched to them; events stay on the Mac