Urgent.News

What's breaking now, across thousands of outlets.

AI

SlothMode AI: Let Local AI Handle the Busywork

This is my submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady. What I Built SlothMode AI is a local AI-powered file organizer designed to make file management less tedious. We all have folders filled with randomly named documents, images, PDFs, and other files. Sorting them manually takes time and effort. SlothMode AI helps automate that busywork by scanning a…

This submission, SlothMode AI, is an open-source AI-powered file organizer aimed at simplifying the tedious task of file management. With countless folders containing randomly named documents, images, PDFs, and other files, manually sorting them is time-consuming and labor-intensive. SlothMode AI automates this process, reducing the effort required and allowing users to focus on more important tasks.

The key features of SlothMode AI include:

- 🤖 AI-powered file categorization using a locally running model

- 📂 Automatic file organization into relevant folders

- 👀 Dry-run mode to preview proposed changes before implementing them

- ↩️ Undo functionality to reverse supported file moves

- 📜 History tracking to review previous organization actions

- 🛡️ Collision handling to avoid overwriting existing files

- 🔌 Heuristic fallback when the local AI model is unavailable

SlothMode AI was developed using Python, Ollama, Qwen2.5 3B, and Streamlit. The application utilizes a locally running language model to propose file categories, while Python handles folder scanning, organization plans, file operations, history, and undo functionality. Streamlit provides a user-friendly web interface for interacting with the tool.

Additional safety features have been incorporated to ensure the automation is trustworthy, such as previewing changes, approving proposed actions, tracking file operations, and supporting undo. The project also includes automated tests for important behaviors, including dry-run safety, collision handling, and undo operations.

Open innovation was crucial in the development of SlothMode AI. By using an open-weight model through local inference, users gain more control over the processing of their files. This approach reduces dependence on external services, avoids per-request API costs, and makes experimentation more accessible. Local processing is especially beneficial for personal files, as users can keep their documents on their own machines, ensuring data privacy and security.

In summary, SlothMode AI aims to minimize the repetitive effort involved in manually sorting files. Users can preview a proposed organization plan, approve changes they wish to implement, and review or undo supported operations as needed. The project also includes a demo folder with sample files for testing the workflow. The use of open-source AI has transformed this repetitive task into a more convenient, user-controlled workflow.

This project demonstrates how open-source AI can simplify routine tasks and enhance user experience.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Give your AI assistant live SEC filings in 5 minutes (MCP tutorial)

TL;DR: one claude mcp add (or a few lines of JSON for Cursor) gives your agent SEC EDGAR filings with real numbers and source links. Steps below, plus the raw API if you'd rather skip MCP.

  • Connect AI tool to Dataville's API for live SEC filings.
  • Server enables real-time access with minimal setup.
  • Ask questions about filings and receive direct answers.

My Android AI agent can only open links it already has a reason to open

I'm building Kryzz AI, a free, open-source (MIT) Android app with an agent mode that can search the web, read pages and write files.

  • Kryzz AI app restricts agent's URL opening to chat links, user-specified sites, or search results.
  • Prevents agent from autonomously navigating to new URLs or creating fake citations.
  • Implements per-turn cost cap ($0.25) to limit agent's running time and prevent bill escalation.

More from Thursday 22 October →