private ai assistant

Private AI Assistant

Keys in Keychain. Encrypted conversations. No GuideLayer server in the loop. A private AI assistant that routes through your own sign-ins.

Every AI assistant privacy policy says the right things. Fewer of them let you verify what leaves your machine.

Screen-aware AI is useful. It's also the category where trust matters most: you're giving software permission to see what you're working on.

GuideLayer is built local-first. Keys in Keychain. Encrypted conversations. No GuideLayer server in the inference loop.

What actually works

Your sign-ins, not ours

Runs on Claude Code and Codex accounts already on your Mac. GuideLayer doesn't host your API keys on a remote server.

Screenshots off by default

Screen capture happens on demand when you ask. Not a continuous feed to the cloud.

Privacy Center

Control memory, screenshots, and routing in one place. One-switch incognito when you need it.

Encrypted at rest

Conversations and saved tutorials encrypted on the device. Not plaintext in a cache folder.

Local-first, not air-gapped

GuideLayer isn't a fully offline model you download once and never phone home. It routes through your Claude Code and Codex sign-ins for inference, with optional metered API fallback. The privacy bet is narrower: no GuideLayer cloud that sees your screen, stores your keys, or sits between you and the model provider you already chose.

Workflow examples

Sensitive client work on screen

Law firm, finance, health admin. The screen has data you can't paste into a random web chat.

  1. 01Enable incognito for the session. Screenshots and memory stay off.
  2. 02Ask by voice. GuideLayer captures context only when triggered.
  3. 03Review Privacy Center settings before and after the task.

Daily use without overthinking it

You want screen help but don't want to re-read a privacy policy every morning.

  1. 01Set defaults once: screenshot on demand, encrypted memory on.
  2. 02Use your existing Claude Code or Codex sign-in.
  3. 03Incognito is there when the work gets sensitive.

Comparing to cloud-only assistants

You're evaluating whether screen-aware AI is worth the trust tradeoff.

  1. 01Check where keys live (Keychain vs vendor cloud).
  2. 02Check when screenshots fire (on demand vs always).
  3. 03Check if there's a vendor server in the loop beyond the model provider.

Private AI isn't a marketing adjective. It's where your keys live, when your screen gets captured, and who can see the conversation besides you.

Common questions

Does GuideLayer upload my screen to a GuideLayer server?

No. Screen capture happens on demand when you ask. GuideLayer routes through your Claude Code or Codex sign-ins — there's no GuideLayer cloud in the inference loop.

Is GuideLayer fully offline?

No. It's local-first, not air-gapped. Inference uses your existing AI sign-ins, with optional metered API fallback. Privacy Center controls memory, screenshots, and incognito sessions.

Where are my API keys stored?

In the macOS Keychain, device-only. GuideLayer doesn't host your credentials on a remote server.

Related guides

Try GuideLayer on your Mac

Screen-aware, private, local-first. Download for Apple Silicon with Claude Code and Codex already signed in.

← GuideLayer homepage