How I Built SlothFlow: A Local-First AI File Organizer That Cleans Your Mess While You Relax

Описание и применение

SlothFlow — локальный AI‑органайзер файлов, который автоматически классифицирует, переименовывает и перемещает документы, сохраняя конфиденциальность и предоставляя безопасный откат.

Задача

Users struggle with disorganized local folders containing PDFs, text, markdown, and other documents, leading to tedious manual renaming and filing, while cloud‑based AI organizers pose privacy risks and subscription costs.

Результат

SlothFlow achieves local‑first AI file organization, eliminating manual effort, preserving privacy, and providing undo capability, all on standard developer laptops without GPU clusters.

Рабочий процесс

• Select a folder or load demo files
• Text extraction via PyMuPDF
• Model inference with local Ollama (Qwen2.5‑1.5B, Gemma 2‑2B, or Llama 3.2‑1B) to produce structured JSON
• Generate organization plan with suggested filenames and target folders
• Preview plan in interactive table (optional inline rename and approval)
• Execute atomic file moves with collision avoidance
• Log actions in SQLite audit trail
• Provide safe undo option

Доказательства

SlothFlow is a local-first web application powered by open-weight AI models (Qwen2.5‑1.5B, Gemma 2‑2B, or Llama 3.2‑1B) running on localhost via Ollama.

Источник

It inspects documents on your machine, determines their categorization, generates human-readable filenames, and safely organizes them into clean category folders— all while ensuring zero document bytes ever leave your laptop.

Источник

Safe Atomic Move & SQLite Audit Log (Instant 1-Click Safe Undo)

Источник