Panda - Hands-Free Assistant

Steepwind Labs

In-app purchases

About this app

Meet Panda - Automate with Voice, your personal AI phone operator. Go beyond standard voice assistants and experience true hands-free control of your device. Panda is an intelligent agent that understands your commands and operates your phone's interface to accomplish complex tasks across any application, just like a human would.

You touch grass. I'll touch your glass.

🤖 What Can Panda Do?
Panda isn't limited to a few pre-programmed actions. By using Android's Accessibility Service, it can see and interact with your screen to perform a vast range of tasks. Simply tell Panda what you want to achieve, and it will handle the taps, swipes, and text input for you.

Examples:

"Send a WhatsApp message to Mom saying 'I'll be home late'."

"Open YouTube and search for 'ambient focus music'."

"Navigate to the last open app and copy the tracking number."

"Check my latest emails and read me the subject lines."

🚀 Key Features
Universal Cross-App Control: Panda is not restricted to specific apps. Its power comes from its ability to navigate and operate any application you have installed, from social media and messaging to productivity and entertainment.

Proactive Automations: Set up powerful workflows that run automatically!

Scheduled Automations: Have Panda perform a task at a specific time (e.g., "Every weekday at 8 AM, check my calendar for today's events and read them to me").

Notification Automations: Begin a task when you receive a notification from a specific app (e.g., "When I get an email from 'Boss', read it out loud").

Intelligent Screen Understanding: Panda uses an advanced semantic parser to understand the elements on your screen—buttons, text, input fields, and more. This allows it to form a plan and execute your commands with precision.

High-Quality Voice: Featuring a natural and clear voice powered by Google Cloud's Chirp models, interactions with Panda are smooth and pleasant.

On-Device File System: For complex tasks, Panda can use a private, on-device file system to save information, create to-do lists, and track its progress, ensuring no step is missed.

♿ Designed for Accessibility
Panda is a game-changer for users with visual and motor impairments. If you rely on accessibility tools but seek a faster, more fluid way to control your device, Panda offers a new level of independence.

True Hands-Free Navigation: Execute complex, multi-step tasks without ever needing to touch the screen.

Voice-Powered Efficiency: Bypass cumbersome screen reader navigation for common tasks.

Full Device Control: Open apps, type messages, change settings, and more, all with the power of your voice.

🔒 Privacy & Permissions
Your privacy is paramount. To function, Panda requires the Accessibility Service permission. Here’s what it means and how we protect you:

Why is it needed? This permission is the "eyes and hands" of the agent. It's what allows Panda to read the UI elements on your screen and perform gestures (taps, swipes) on your behalf. Without it, cross-app automation is impossible.

Your Data: Screen content and conversations are sent to Google's Gemini AI models for the sole purpose of understanding and executing your command in real-time. We are committed to transparency, and you can review our full Privacy Policy for details.

You Are in Control: The Accessibility Service is only active when you give Panda a task. You can revoke this permission at any time in your phone's settings.

Ratings and reviews

Still to early for prime time. It's unresponsive. Very very very slow when it does work. I just did a basic task to send a message took forever almost 4 mins to do anything. Then I top of that it's always popping to listen which is annoying. Because when you want to use it. It doesn't work. App is not ready for general use. Cool idea I hope it works out for them

This app is well-designed for an indie project, if overt with the art (2013 design language). Functionally, though, I can't recommend this app for power users because BYOK is paywalled (0.70€/mo as of writing), meanwhile AI Studio offers keys for free. Let's say it wasn't: there's no voice input (offloaded, ignores LLM's own), TTS is forced, the agent is fully reliant on the a11y service - no shell cmds, no MCP. It's recompilable, but I'd rather semaphore through Gemini in Calendar.

voice input doesn't work at all, as soon as I tap micro, immediately, it says "please repeat" and disables voice input. all permissions are given.

What’s new

Automation Fix