Your mind will never go blank in a call again

They ask the hard question. You already have the answer.

Interviews aren't fair — so level the playing field. An AI assistant that sits on your screen during any call, invisible to everyone but you. Real-time answers when you need them most. No one sees it. No one knows.

Get Lifetime Access — €79 See How It Works
macOS Windows (soon) Linux (soon)
Zoom Meeting — Interview with Google
00:23:47
2
E2E
You
Self view
Sarah Chen
Hiring Manager
Only visible to you — invisible to screen share
Ask me anything...
AI Response
⌘K

System: "How would you design a distributed cache?"

Use a multi-layered approach: client-side cache (Caffeine) → distributed layer (Redis with consistent hashing) → write-behind with async replication for durability.

Introducing Glass Cloak

Meet your invisible AI advantage

See why hundreds of candidates trust Glass Cloak to never freeze in interviews again.

Watch the intro

Hundreds of people already use Glass Cloak daily
0px
Visible to Screen Recorders
<1s
Real-Time Answer Speed
0%
Undetectable on Screen Share
Zero
Network Trace With Local AI
Stealth Features

Built to be invisible.
So you never freeze again.

That moment when your mind goes blank mid-interview? Gone. Get real-time answers during any call — without anyone knowing you have help.

Screen-Recording Proof

Glass Cloak is invisible to screen recordings and screen shares. Content protection is baked into the native window layer — not a CSS hack, not an overlay.

  • Invisible on Zoom, Google Meet, Teams, and all screen recorders
  • Native contentProtected: true — OS-level
  • Screen share captures blank space — nothing is leaked
What you see
AI: Use consistent hashing with ring topology...
What they see
Nothing here.

Always On Screen. Never Detected.

A razor-thin 600px floating bar at StatusWindowLevel. Sits above every app on every desktop. Doesn't steal focus. Doesn't flash.

Real-Time Transcription

Hear a question, see it transcribed instantly. Local Whisper.cpp on-device or cloud Whisper API. Every word captured as it's spoken.

Hidden From Your Dock

Removes itself from dock, taskbar, and app switcher via ActivationPolicy::Accessory. Invisible cursor. Zero visual trace.

Glass Cloak is not here

Local AI. Zero Network Trace.

Built with Rust for native performance. When using built-in models, there is zero internet traffic — no API calls, no telemetry, no logs. Your questions and answers exist only on your machine.

  • Qwen 2.5 3B LLM runs locally via llama.cpp
  • Whisper STT runs locally via whisper.cpp
  • Metal acceleration on Apple Silicon, CUDA on NVIDIA
Network Activity Monitor
TCP Connections:0
HTTP Requests:0
DNS Lookups:0
Data Sent:0 bytes
Data Received:0 bytes
Local Model:qwen2.5-3b-instruct-q4_k_m.gguf
Local STT:whisper-base.bin
Status:All local. No trace.
100% Undetectable

Works everywhere
you have calls

Zoom, Meet, Teams, HackerRank — doesn't matter. We test daily on every platform. Nobody will ever know you're using it.

Zoom
Undetectable
Google Meet
Undetectable
Microsoft Teams
Undetectable
Cisco Webex
Undetectable
Amazon Chime
Undetectable
HackerRank
Undetectable
CoderPad
Undetectable
LeetCode
Undetectable
Why Glass Cloak

Other tools leave you exposed.
We don't.

Most AI tools charge you monthly, send your data to the cloud, and can get you caught. We built something that actually protects you.

Typical AI Assistants

How most tools work
  • Monthly subscription required
    Recurring payments every month. Cancel and you lose access immediately. Costs add up over time.
  • No local AI models
    100% dependent on cloud APIs. No option to run AI locally on your own hardware.
  • Cloud-only processing
    Every prompt and response goes through external servers. No way to keep things fully on-device.
  • Single platform or browser-only
    Often limited to one OS or run as a browser extension. Not a native desktop app.
Monthly cost that adds up
No local AI, no lifetime option

Glass Cloak

What makes us different
  • €79 once — lifetime access
    One payment. No subscriptions. No renewals. Every future update included. You own it.
  • Built-in local AI models
    Qwen 2.5 3B (llama.cpp) + Whisper (whisper.cpp) run directly on your hardware. Use AI without cloud APIs when you want to.
  • Local + Cloud — your choice
    Switch between local models and GPT-4o / Claude anytime. Use your own API keys — no middleman taking a cut.
  • Native app on macOS, Windows & Linux
    Built with Rust and Tauri. Not a browser extension. Lightweight, fast, and designed for all three platforms.
One-time payment
€79 lifetime
local AI + cloud AI + all platforms
Get Lifetime Access — €79

Stay ahead in every call.
Pick how much power you need.

Want zero trace? Use local AI — nothing leaves your machine. Need the best answers for a tough interview? Switch to GPT-4o or Claude in one click.

Stealth Mode

Offline
  • Qwen 2.5 3B — local LLM via llama.cpp
  • Whisper — local STT via whisper.cpp
  • Zero network traffic — nothing to detect
  • No API calls, no logs, no trace

Maximum Power

Your Keys
  • OpenAI — GPT-4o, GPT-4, Whisper API
  • Anthropic — Claude Sonnet, Claude Opus
  • Your own keys — no middleman, no data retention
  • Switch between stealth and power mode instantly
Get Started

Ready in under 2 minutes

Download. Activate. Walk into your next interview knowing you'll never blank out again.

1

Download

One lightweight installer for macOS, Windows, or Linux. No browser extensions. No meeting bots. No sign-up required.

2

Activate & Go Stealth

Enter your access key. Download local AI models. Glass Cloak auto-hides from your dock and enables content protection — stealth mode is on by default.

3

Never freeze again

Tough question? The answer is already on your screen. Real-time transcription captures what they say. AI gives you the perfect response. Nobody sees a thing.

Limited Time Offer

One price. Yours forever.

Everyone else charges monthly. We don't. Pay once and walk into every call with confidence for the rest of your career.

LIFETIME DEAL — LIMITED TIME
Lifetime Access
€79
67% OFF — Introductory Price

One-time payment. Lifetime updates. No recurring fees. Ever.

  • Screen-recording proof — invisible on all platforms
  • Hidden from dock, taskbar & app switcher
  • Invisible cursor mode
  • Local AI — Qwen 2.5 3B + Whisper (on-device)
  • Cloud AI — GPT-4o, Claude (bring your own keys)
  • Real-time transcription & instant answers
  • macOS, Windows & Linux
  • Lifetime updates & priority support
  • Commercial use license
Get Lifetime Access — €79

Secure checkout. Instant access. Download immediately after purchase.

Frequently asked questions

Yes. Glass Cloak uses native OS-level content protection (contentProtected: true on macOS, equivalent on Windows/Linux). This means the window is excluded from screen capture at the operating system level — not through CSS hacks or overlays. Screen recorders, Zoom, Google Meet, Teams, and any other screen sharing tool will only see a blank space where Glass Cloak sits.

Glass Cloak hides from the dock, taskbar, and app switcher using ActivationPolicy::Accessory. It also supports invisible cursor mode. When using local AI models, there is zero network traffic — no API calls, no DNS lookups, nothing. Even DLP (Data Loss Prevention) software monitoring network activity will see nothing.

Glass Cloak ships with Qwen 2.5 3B for language generation (via llama.cpp) and Whisper for speech-to-text (via whisper.cpp). Both models run entirely on your machine using your GPU (Metal on Apple Silicon, CUDA on NVIDIA). No internet connection required. No data ever leaves your computer.

Absolutely. You can add your own OpenAI or Anthropic API keys in Settings. When you need the most powerful reasoning for complex technical questions, switch to cloud models. When stealth is critical, switch back to local models instantly.

macOS 13+ (Apple Silicon & Intel), Windows 10+, and Ubuntu 22.04+. It's built with Tauri and Rust for native performance on every platform — no Electron bloat.

Pay €79 once and get Glass Cloak forever. No monthly fees. No annual renewals. All future updates and new features included. This is a limited-time introductory price that will increase.

When using local models, your data never leaves your machine — period. Zero telemetry, zero tracking, zero analytics. When using cloud APIs, requests go directly from your machine to OpenAI/Anthropic using your own keys — we are never a middleman.

Stop going in unprepared.
Start showing up ready.

The best candidates aren't smarter — they're better prepared. Invisible AI on your screen. Answers when you need them. No one will ever know.

Download for
macOS (Apple Silicon)
Download for
macOS (Intel)
Coming soon
Windows
Coming soon
Linux

Latest version: v0.0.35

macOS 13+ (Apple Silicon & Intel). Windows & Linux coming soon. No meeting bots. No browser extensions.

Never blank out in a call again — €79 for life

Get Lifetime Access — €79