ScreenStash: Turn Screenshots into Actionable Plans with Local Gemma 3 & Render A developer built ScreenStash, an open-source Tauri 2 desktop app for Windows and macOS that watches a screenshot folder, extracts text via OCR, and uses a local Gemma 3:4b model through Ollama to turn screenshots into structured tasks, events, receipts and references. The app stores results in a local SQLite database, restricts Ollama calls to loopback endpoints so screenshot text cannot be sent to a remote server, and offers an optional FastAPI cloud runtime on Render for users who opt into cloud processing. The MIT-licensed codebase pairs a React 19/TypeScript frontend with a Rust backend and ships via a GitHub Actions release pipeline. We all do it: we hit Win + Shift + S or Cmd + Shift + 4 a dozen times a day. ~/Pictures/Screenshots Screenshot 482 .png , and the tasks or meetings you captured are long forgotten. .pyw detects new screenshots with zero user intervention and triggers a floating OS toast notification. Show in Folder . ✦ AI: Semantic concept match expense: "receipt" · Direct match in title explaining exactly why the AI selected it. --- Code ScreenStash is a downloadable Tauri desktop application for Windows and macOS. It watches a user-selected screenshot folder, extracts text locally, and stores organized results in a local SQLite database. Download the .exe or .dmg from the landing page or GitHub Releases. The distributed application does not require Python or a source checkout. ScreenStash accepts only loopback Ollama endpoints localhost , 127.0.0.1 , or ::1 so screenshot text cannot be sent to a remote Ollama server by configuration. The React/Tauri UI lives in desktop/ ; Python is packaged only as the processing sidecar. cd desktop npm ci … The codebase is 100% open-source under the MIT license, structured into: desktop/ : Tauri 2 desktop shell, React 19, TypeScript, and custom CSS design system. desktop/src/aiFilter.ts : Dual-tier semantic intent engine and Ollama JSON inference client. desktop/src/Calendar.tsx : Dynamic 7-day monthly calendar and temporal date extractor. desktop/src-tauri/ : Rust backend implementing native IPC, asset protocols, file deletion, and OS shell commands. cloud service/ : FastAPI cloud runtime deployed on Render. .github/workflows/ : Automated GitHub Actions release pipeline building Windows MSI and NSIS packages. ScreenStash follows a privacy-first, layered architecture that transforms raw screenshots into searchable, actionable information. 01 → Screenshot Capture Win + Shift + S . 02 → Duplicate Detection SHA-256 hash to identify duplicate screenshots. 03 → Text Extraction 04 → AI Processing Extracted text is processed through a flexible, multi-tier AI pipeline: | 💻 Local AI | ☁️ Cloud AI | |---|---| | Gemma 3:4b + Ollama | Render + FastAPI | | Runs inference locally using structured JSON output. | Provides an alternative for users who opt for cloud processing. | | Keeps inference on-device. | Uses the hosted cloud runtime. | 05 → Local Data Storage 06 → Desktop Experience Powered by Tauri 2 + React , ScreenStash turns processed screenshots into an interactive workspace: We chose Gemma 3:4b as our flagship on-device model. Using Ollama's structured JSON format, we pass the OCR extracted text into Gemma with strict schema constraints to reliably extract: Task , Event , Receipt , Reference meeting , expense , urgent , task , doc . ✦ AI Filter , queries local Ollama or Render Cloud with candidate items in strict JSON mode. evaluateNeedsAttention to score urgency tokens evaluateSavedForLater to detect reference documentation and expense archives. 3. Native Desktop WebView via Tauri 2 & Rust Rather than packaging a bloated 250MB Electron binary, ScreenStash is built on Tauri 2: convertFileSrc safely bridge local filesystem images into the webview without CSP violations. explorer.exe /select,