kellyvv

kellyvv / PhoneClaw

Public

PhoneClaw: On-device AI Agent for Phone powered by Gemma 4

92
11
100% credibility
Found Apr 06, 2026 at 92 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Swift
AI Summary

PhoneClaw is an offline iPhone app running a local AI agent for privacy-focused tasks like calendar management, reminders, contacts, clipboard handling, device info, text tools, and image understanding using on-device Gemma models.

How It Works

1
📱 Discover PhoneClaw

You hear about a private AI helper that runs completely on your iPhone, keeping all your chats and photos safe without internet.

2
🖥️ Put the app on your phone

Connect your iPhone to your Mac and follow simple steps to install the app, like adding a fun new game.

3
🧠 Pick your AI brain

Open the app, choose a smart brain size that fits your phone, and let it download right there—no waiting for big files.

4
🔑 Allow helpful access

Turn on permissions so it can check your calendar, reminders, or contacts when you ask.

5
💬 Start chatting

Ask it anything like 'What's my phone info?' or 'Remind me at 8pm' and watch it help instantly.

✨ Your private assistant is ready

Now your iPhone handles daily tasks like events, reminders, and photo questions all offline and securely.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 92 to 92 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is PhoneClaw?

PhoneClaw is an on-device AI agent for iPhone, powered by Gemma 4 models running entirely locally in Swift. It lets you chat naturally to handle phone tasks like adding calendar events, setting reminders, saving contacts, querying device info, or understanding photos—all without internet or data uploads. Privacy-first design ensures your chats and images stay on-device.

Why is it gaining traction?

It stands out with a file-based skills system using simple Markdown files, letting anyone add or tweak capabilities without recompiling the app. Multimodal image analysis works offline, and flexible model management (E2B for stability, E4B for power) fits iPhone hardware limits. Developers dig the zero-cloud hook for truly private iPhone claw AI.

Who should use this?

iOS devs prototyping personal automation agents, like voice-free Shortcuts alternatives for calendar/reminders. Privacy hawks wanting a local phoneclaw on their A16+ iPhone for daily tasks. Early adopters testing Gemma on-device before building custom phoneclaw apk variants.

Verdict

Grab it if you're into Swift on-device agents—quick setup via Xcode, solid core skills, and easy extensibility make it worth a spin on iPhone 15 Pro+. Low 1.0% credibility score and 92 stars signal early days (basic docs, no tests visible), so expect tweaks; fork and contribute for production use.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.