'''# Git Detox

Git Detox is your Git safety net. It's a tool that visualizes local Git activity (branches, stashes, reflog, dangling commits) in a unified view and highlights recoverable or at-risk work.

Features

• Scan for Recoverable Work: Detects deleted branches, dangling commits, and dropped stashes.
• Risk Detection: Highlights work at risk of being lost to Git's garbage collection.
• Hygiene Analysis: Identifies merged and stale branches to keep your repo clean.
• Visual TUI: Interactive terminal interface for exploring and restoring your Git history.
• Safe Recovery: Preview commands before execution with guided restoration.

Installation

# Clone the repository
git clone https://github.com/youruser/git-detox.git
cd git-detox

# Install in editable mode
pip install -e .

Usage

CLI Commands

• git-detox scan: List all recoverable items in the current repository.
• git-detox show <ID>: View detailed information and diff summary for an item.
• git-detox restore <ID>: Safely restore an item to a new branch.
• git-detox health: Get a hygiene report and health score for your repo.
• git-detox tui: Launch the interactive terminal UI.

Example

$ git-detox scan
Scanning repository...
                              Recoverable Git Work                              
┏━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━┓
┃ ID          ┃ Type           ┃ Source/Ref ┃ Subject         ┃ Risk   ┃ Date           ┃
┡━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━┩
│ del_abc1234 │ deleted_branch │ feature-x  │ Fix login bug   │ low    │ 2026-02-13     │
└─────────────┴────────────────┴────────────┴─────────────────┴────────┴────────────────┘

Strategic Advice for Users

1. Don't Panic: If you just deleted a branch or reset your HEAD, git-detox can likely find it in the reflog.
2. Check Health Regularly: Use git-detox health to keep your local environment lean.
3. Preview Before Restore: Always use show to verify the content before restoring.

License

MIT '''