ayushh0110

ayushh0110 / ScreenMind

Public

🧠 AI-powered screen memory — captures, analyzes, and lets you search/chat your screen history. Powered by Gemma 4 . 100% local, 100% private.

ai fastapi gemma llama-cpp localai
16
1
85% credibility
Found May 24, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

ScreenMind is a privacy-first AI application that continuously captures screenshots of your screen, analyzes them using a local AI model (Gemma 4), and builds a searchable memory of your daily screen activity. It runs entirely on your machine with no cloud dependencies, letting you chat with your screen history, record voice memos, transcribe meetings, and run automated agents — all while keeping your data private through features like sensitive data filtering, screenshot encryption, and an optional dashboard lock PIN.

How It Works

1
💡 You hear about ScreenMind

Someone tells you about an app that watches your screen and remembers everything you've done — but keeps it all on YOUR computer, private and secure.

2
🖥️ You install and launch it

You run one simple command, and ScreenMind starts up on your machine. It downloads the AI model once, then everything runs locally forever.

3
🧠 It quietly watches your screen

ScreenMind runs in the background, capturing screenshots only when your screen actually changes. It analyzes each one with a local AI that understands what you're doing.

4
🔍 You search your screen history

You ask questions like 'what did I work on yesterday at 2pm?' or 'find that Discord message about the project' and get instant, specific answers.

5
You explore in different ways
🗣️
Voice memos

Hold a hotkey, speak your thought, and it captures a screenshot plus transcription right at that moment.

🎙️
Meeting notes

It auto-detects when you're in a Zoom or Teams call, records audio, and transcribes everything.

🔒 You have a complete, private AI memory

Everything — screenshots, transcripts, summaries — stays on your machine. No cloud, no tracking, no telemetry. Your digital life, organized and searchable, entirely under your control.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 16 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 ScreenMind?

ScreenMind is a local AI tool that continuously captures your screen, analyzes it with Gemma 4, and lets you search or chat with your entire visual history. Think of it as a searchable memory of everything you saw and did on your computer. Built in Python, it runs entirely on your machine with no cloud dependencies, storing screenshots and AI-generated summaries in a local SQLite database with full-text search.

Why is it gaining traction?

The privacy angle is the hook. Microsoft Recall stored screen data in plaintext and sent telemetry, sparking massive backlash. ScreenMind positions itself as the open-source, privacy-first alternative that keeps everything local. Beyond that, the conversational chat interface is genuinely useful—you can ask "what did Ishaa say on Discord yesterday?" and get an actual answer. The MCP server integration means your AI coding tools (Claude, Cursor, VS Code) can query your screen history directly. Three analysis modes let you trade speed for depth, and features like voice memos, meeting transcription, and an agent platform for automations make it more than a simple screen recorder.

Who should use this?

Developers who want to remember where they saw something three days ago will get the most value. Power users who care about privacy and want AI capabilities without sending data to the cloud are the target audience. Anyone who attends lots of video meetings and wants searchable transcripts will find it useful. If you need GDPR-compliant screen logging or want to build automations on top of your activity history, this is worth evaluating.

Verdict

ScreenMind is a well-architected project with a compelling privacy story and solid feature depth, but the 16-star count signals early-stage software that needs real-world hardening. The 0.85% credibility score reflects a young project with limited community validation. Try it if you want local-first screen memory, but expect to encounter rough edges and contribute fixes back.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.