smart-mv 0.2.0

Smart Move (smv) - AI-powered file organization tool

smart-mv

smart-mv (smv) is an AI-powered file organizer that analyzes file content and suggests where to move files, how to rename them, and when to trash clutter.

Highlights

• Multi-provider AI setup with persistent profile:
  • Ollama (first in selection list)
  • OpenAI
  • Anthropic
  • Google Gemini
• Model selection per provider (with Ollama local model discovery).
• API key persistence with OS keyring and explicit plaintext fallback.
• Faster sorting flow with deterministic trash shortcuts and depth-limited search.
• Provider/model settings can be changed anytime from CLI.
• Richer document understanding with native .docx text extraction.
• Filesystem tool-calling (OpenAI/Ollama) so AI can list directories and find files/folders during destination planning.

Quick start (uv-first)

Install

uv tool install smart-mv

or from source:

git clone https://github.com/kariszhuang/smart-mv.git
cd smart-mv
uv sync
uv run smv --version

Configure AI once

uv run smv ai setup

This interactive setup lets you choose provider, model, base URL (where applicable), and API key storage.

Organize a file

uv run smv /path/to/file.pdf

AI configuration commands

# Show current profile
uv run smv ai show

# Interactive reconfiguration
uv run smv ai setup

# Set provider/model/base URL directly
uv run smv ai set-provider ollama
uv run smv ai set-model gemma3:12b
uv run smv ai set-base-url http://localhost:11434/v1

# Set API key (keyring by default; falls back to plaintext if needed)
uv run smv ai set-api-key --storage keyring

# List suggested/discovered models
uv run smv ai list-models --provider ollama

Development

uv sync
uv run python -m unittest discover -s tests -q

Python version

This project now targets Python 3.10+.