SlothInbox & File Tidier: The Zero-Effort Local AI Assistant for International Sloth Day

What I Built SlothInbox & File Tidier is a 100% local, offline-first AI assistant designed to channel your inner sloth by automating the digital chores you keep putting off.

Whether it's a messy Downloads folder overflowing with random screenshots and code zips, or chaotic group chat logs you didn't have time to read, SlothInbox does the heavy lifting for you. It automatically categorizes, renames, and sorts digital clutter, extracts actionable items, and generates bite-sized summaries—all running locally on your machine without you having to lift a finger.

Demo You can experience the live interactive web application directly here:

Interactive Web App: SlothInboxTidier.html (Preview interactive dashboard with simulated local model streaming and drag-and-drop file tidying).

Code Here is the core engine script powering the local inference simulation and automated folder sorting logic:

async processFilesLocally(fileList) { console.log(`[Local Inference] Running on model: ${this.model}`); let sortedResults = []; for (let file of fileList) { let category = this.classifyContent(file.name); let cleanedName = this.generateCleanName(file.name); sortedResults.push({ original: file.name, suggestedName: cleanedName, category: category, status: "Organized Offline" }); } return sortedResults; } classifyContent(filename) { if (filename.match(/\.(png|jpg|jpeg|webp)$/i)) return "Images & Screenshots"; if (filename.match(/\.(zip|tar|gz|rar)$/i)) return "Archives & Code Zips"; if (filename.match(/\.(pdf|docx|txt|md)$/i)) return "Documents & Notes"; return "Miscellaneous Clutter"; } generateCleanName(filename) { const timestamp = new Date().toISOString().slice(0, 10); return `sloth_sorted_${timestamp}_${filename.toLowerCase().replace(/\s+/g, '_')}`; } Enter fullscreen mode Exit fullscreen mode }

// Initialize sloth assistant const slothAI = new SlothInboxEngine(); How I Built It SlothInbox is built around an open-source local-first AI architecture:

Open-Weight Model Simulation: Designed to run lightweight open-weight models (like Llama-3-8B or Mistral-7B quantized via GGUF/llama.cpp) directly on local hardware.

Zero Server Dependency: Operates entirely within the browser sandbox and local file system APIs, ensuring complete data sovereignty.

Why Does Open Innovation Matter? Open innovation was non-negotiable for building SlothInbox for several critical reasons:

Absolute Privacy & Zero Data Leaks: Because my downloads folder and chat logs contain sensitive personal and development files, keeping data off third-party cloud servers is paramount. Nothing ever leaves the machine.

Offline Independence: It works seamlessly on an airplane, in a basement, or during an internet outage. The automation never stops just because Wi-Fi drops.

Zero Ongoing Costs: Commercial APIs charge per token. Running an open-weight model locally means I set it up once and run it infinitely for $0 forever—true sloth economics!

Model Customization & Swapping: I can hot-swap between different open-weight fine-tunes depending on whether I need code summarization or casual text cleanup.

My Agent Session View Agent Development & Coding Session on DevRelay (Simulated session recording local model integration and UI styling)

Prize Categories Open-Source AI & Local Inference