VisionVault — Offline Movie & TV Library Manager

VisionVault is a modern offline movie & TV library manager built with CustomTkinter, Flask, and SQLite.
Manage your personal collection, browse cinematic poster grids, track watch history, launch local files, and access your library through the built-in Web UI / TV Mode for browser and TV playback.

No accounts • No cloud • No subscriptions • 100% local

Optional Wikipedia-powered metadata and posters make it easy to enrich your collection while keeping everything private and under your control.

A personal media vault designed for collectors who want a clean desktop experience with the flexibility of local web playback.

---

Key Features

Library Management

• Add movies by title or file
• Automatic title and year detection from filenames
• Local SQLite database storage (movies.db)
• Edit full metadata:
  • Title
  • Year
  • Genres
  • Runtime
  • Resolution
  • Overview
  • Poster
  • File path
• Delete entries cleanly
• Track watch history
• Launch movies directly from the desktop app

---

TV Show Support

• Import entire TV show folders
• Automatic show and episode structure
• Episode naming format:
  • S01E01 - Episode Name
  • S01E02 - Episode Name
• Manual editing of:
  • Episode posters
  • Descriptions
  • Metadata
• Progress tracking per show
• Automatic show poster inheritance from episode artwork
• Optional poster propagation across episodes that do not already have artwork

---

TV Mode Web Interface

• Launch a browser-based VisionVault TV interface directly from the desktop app
• Browse your library from another device on your local network
• Stream movies directly in-browser
• TV-friendly navigation layout for couch or remote-style use
• Dedicated sections for:
  • Continue Watching
  • Movies
  • Shows
  • Recently Added
• Built-in movie detail pages, stats page, and full movie library page

---

Resume Playback

• Browser playback now saves watch progress
• Resume movies from your last saved position
• Resume and Start Over options appear on supported items
• Completed playback clears saved progress automatically
• Browser playback and desktop stats stay in sync

---

Grid View (Modern Media Wall)

• Poster-focused tiled layout
• Uniform sizing with clean alignment
• Clamped and wrapped titles without stretching
• Dense, directory-style browsing
• Fully integrated with:
  • Sorting
  • Filters
  • Show navigation
  • Right-click context menu
  • Keyboard navigation

---

List View

• Clean detailed list layout
• Highlighted selection
• Larger readable text spacing
• Fast keyboard-friendly browsing

---

Search, Filter & Sort

• Live search by title
• Filter by:
  • All
  • Watched
  • Unwatched
• Genre filter (auto-generated)
• Sort by:
  • Title A-Z / Z-A
  • Year up / down
  • Watched up / down
  • Recently added / oldest added

---

Stats Dashboard

• Total movies, shows, and episodes
• Watched vs unwatched breakdown
• Total watch counts
• Top watched movies and episodes
• Most common genres
• Recently added items

---

Keyboard Navigation

• Full keyboard navigation support in the desktop app
• Shortcuts include:
  • Enter to play
  • Space to mark watched
  • E to edit
  • Delete to remove selected item
  • Backspace / Escape to return from show episode view
• Smooth auto-scroll keeps selected items visible while browsing

---

Themes and Interface Customization

• Multiple accent color themes
• Supports Dark / Light / System appearance modes
• Theme styling updates buttons, highlights, borders, and dialog accents
• Cleaner desktop-style menu bar for app controls

---

Quality of Life

VisionVault automatically remembers:

• Window size and position
• Split panel layout
• Dark / Light / System appearance
• Active theme
• Last selected item
• Grid or List view mode

Saved locally in:

movie_inventory_settings.json

Posters are cached locally in:

posters/

---

Getting Started

Add by Title

1. Type into Add by Title...
2. Click Add
3. Optionally fetch:
   • Wikipedia metadata
   • Poster art
4. Save

---

Add by File

1. Click Add by File
2. Select a video file
3. VisionVault will:
   • Guess title and year
   • Detect runtime and resolution using ffprobe if installed
4. Edit if needed and save

---

Import TV Shows

1. Click Import TV Show
2. Select a folder of encoded episode files
3. VisionVault will build the show and episode structure automatically
4. Edit posters or metadata if needed

---

Watching and Playing

• Play opens your movie in the default OS player from the desktop app
• Successful desktop playback increments watch count automatically
• In TV Mode, movies can be streamed directly in-browser
• Browser playback supports saved resume progress
• You can also manually use Mark as Watched

---

TV Mode

1. Open the TV Mode menu from the desktop app
2. Start the TV web interface
3. Open the provided local network URL on your browser or TV
4. Browse and stream your collection from the web interface

---

Notes

• Internet is only required for Wikipedia metadata fetching
• The desktop app works fully offline otherwise
• TV Mode works over your local network
• All posters are stored locally for fast loading
• If you move files, update the file path in Edit Details

---

Built With

• CustomTkinter
• Flask
• SQLite
• Wikipedia Action API

---

Download

Grab the standalone Windows .exe from the Releases page when a new version becomes available.

Manual Install

1. Clone the repository

git clone https://github.com/VisionDepth/VisionVault.git
cd VisionVault

2. Create and activate a Conda environment

conda create -n VisionVault python=3.13
conda activate VisionVault

3. Or create a standard virtual environment

python -m venv venv

4. Install dependencies

pip install -r requirements.txt

5. Launch VisionVault

python visionvault.py

Notes for Manual Install

• vault_core.py, visionvault.py, and visionvault_web.py should stay in the same project folder
• TV Mode uses the same local database as the desktop app
• Posters are stored locally in the posters/ folder
• Settings are stored in movie_inventory_settings.json
• ffprobe is recommended for automatic runtime and resolution detection