downloads-organizer 1.0.3

Smart downloads folder organizer powered by local AI (Ollama)

Downloads Organizer

Smart downloads folder organizer powered by local AI (Ollama). Compatible with macOS and Linux. Installs with pipx and runs as a system service.

Your files never leave your machine — all AI classification happens locally via Ollama.

---

Demo

~/Downloads/
├── 📥 Recién Descargado/      ← files from today
└── 📁 Organizado/
    ├── 🔴 Urgente/
    │   ├── Trabajo/
    │   │   └── documento/
    │   └── Finanzas/
    │       └── hoja_de_calculo/
    ├── 🟡 Normal/
    │   └── Personal/
    │       └── imagen/
    └── 📦 Archivo/

---

Requirements

• Python ≥ 3.10
• pipx — to install the CLI
• Ollama — local AI engine

---

Installation

macOS

# 1. Install Ollama
brew install ollama
ollama serve &
ollama pull llama3.2

# 2. Install pipx (if needed)
brew install pipx
pipx ensurepath

# 3. Install Downloads Organizer
pipx install downloads-organizer

# 4. Enable system service
dorg install

Linux

# 1. Install Ollama
curl -fsSL https://ollama.ai/install.sh | sh
ollama pull llama3.2

# 2. Install pipx (if needed)
pip install pipx --user && pipx ensurepath

# 3. Install Downloads Organizer
pipx install downloads-organizer

# 4. Enable systemd service
dorg install

---

Usage

Command	Description
dorg install	Install and enable system service
dorg uninstall	Remove system service
dorg status	Service status + statistics
dorg classify	Classify pending files now
dorg watch	Start in foreground (debug mode)
dorg logs	View service logs
dorg logs -n 200	View last N log lines
dorg config	Edit configuration file
dorg config --set ollama.model=mistral	Change Ollama model
dorg projects	List approved projects
dorg projects --add Trabajo	Add project manually

---

Configuration

File: ~/.config/downloads-organizer/config.yaml

downloads_folder: ~/Downloads
recent_folder_name: Recién Descargado   # staging folder for new files
organized_folder_name: Organizado        # destination for classified files
classify_time: "06:00"                   # daily classification time

ollama:
  base_url: http://localhost:11434
  model: llama3.2
  timeout: 60
  confidence_threshold: 0.65             # below this → ask user

projects:                                # approved project names
  - Trabajo
  - Personal
  - Finanzas
  - Estudio

---

Recommended Ollama Models

Model	Size	Speed	Quality	Best for
llama3.2	2 GB	Fast	Good	Default, everyday use
mistral	4 GB	Medium	Very good	More nuanced classification
phi3	2 GB	Fast	Good	Low-RAM machines
gemma2	5 GB	Slow	Excellent	Best accuracy

Change model:

ollama pull mistral
dorg config --set ollama.model=mistral

---

How It Works

New file detected
      │
      ▼
[watchdog event]
      │
      ▼
_should_move()? ──No──► ignore
      │
     Yes
      │
      ▼
wait for file ready (download complete)
      │
      ▼
move → Recién Descargado/
+ native notification
      │
      ▼
      (daily at 06:00)
      │
      ▼
[Ollama classification]
      │
      ├─ confidence ≥ 0.65 + known project
      │         └─► move automatically
      │
      ├─ confidence < 0.65
      │         └─► show interactive dialog
      │
      └─ new project suggested
                └─► ask user to approve/rename/reject
                          └─► move to Organizado/

---

Contributing

See CONTRIBUTING.md for development setup, commit conventions, and the pull request process.

---

License

MIT — see LICENSE. Copyright (c) 2025 Jarsa Sistemas