Listweavarr

Weave your next watch. Curated lists, connected stories, and a library worth exploring.

Listweavarr is a single-user, self-hosted movie and TV discovery workbench built around lists and watch orders. Import a franchise timeline, match movies/seasons/episodes, compare it with Plex, discover what is missing, and create an ordered Plex playlist. Requests to Radarr and Sonarr are explicitly reviewed.

What it does

• Import supported Trakt, Letterboxd, MDBList, IMDb and ChronoCanon lists, files, text, articles and public Google Sheets.
• Review inferred matches before saving; correct IDs, episode ranges, anime numbering and mixed movie/TV sequences.
• Keep local collections or follow linked sources; curate multiple watch orders and share portable list files.
• Explore recommendations informed by your Plex library, personal ratings and watch history, with ignore, language and release filters.
• Track upcoming releases, creator projects, requested downloads and availability notices.
• Build Favorites from Plex, plan Tonight sessions and find franchise/library gaps.
• Create Plex playlists with independent square covers. Preview Official catalogue updates separately from personal-copy and Plex updates.

Public community hosting is not deployed. File sharing works today. Curator mode prepares local Official versions; actual shared publishing requires a separately managed repository/service.

Run locally

Requires Python 3.11+. No pip packages, Node or database server are needed at runtime.

git clone https://github.com/RayRayElite/Listweavarr.git
cd Listweavarr
python app.py

Open http://127.0.0.1:8765. Windows users can double-click Start Listweavarr.cmd. Keep the process running; Ctrl+C stops it. Settings, credentials, lists and caches are stored in data/, which is excluded from Git.

Connect Plex and TMDB in Settings ? Connections. Optional integrations include Radarr, Sonarr, Trakt, MDBList, TheTVDB, Prowlarr and Discord. Use API credentials issued to you; no shared credentials are bundled. Some discovery and metadata features require their provider connection.

Run persistently with Docker

Requires Docker Engine/Desktop with Compose v2. From the repository directory:

docker compose up -d --build
docker compose logs -f

Open http://127.0.0.1:8765. The named volume listweavarr_config persists settings, caches, lists and installed catalogue updates at /config. The container runs as UID/GID 10001:10001, has a read-only application filesystem and restarts unless stopped.

# Update after reviewing the changelog and backing up the config volume:
git pull --ff-only
docker compose up -d --build
# Stop without deleting saved data:
docker compose down

Do not use docker compose down -v unless you intend to delete the saved volume. This release builds its image from source; it does not require a Docker registry account.

For NAS/LAN deployment, custom ports, migration from a local installation and backups, see Deployment. Container connections to Plex/Radarr/Sonarr need reachable NAS/service addresses; localhost inside the container means the container itself.

Access and privacy

Listweavarr is single-user, with optional owner sign-in. Local access remains the default until LISTWEAVARR_OWNER_PASSWORD_FILE is configured. Default local and Compose configurations expose it only to localhost. For remote use, configure owner access and HTTPS through an appropriate reverse proxy or use a restricted private network. Host allowlisting and CSRF tokens are browser-request protections, not user authentication.

Credentials stay in the server-side state file and are omitted from ordinary browser state responses and portable backups. Protect the data directory and volume: it contains credentials in plaintext. Do not publish it, an exported credential backup, logs containing provider responses, or screenshots of your private library.

Artwork and metadata

Third-party posters and Kometa/logo collections are not bundled without redistribution permission. Users may add their own images. Original SVG status ribbons are included; provider artwork is loaded from TMDB when configured. Movie/TV names and artwork remain the property of their respective owners.

Metadata and ratings: TMDB, with optional provider integrations. This product uses the TMDB API but is not endorsed or certified by TMDB. See Third-party notices.

Documentation and contributions

• User guide ? Deployment ? Changelog
• Contributing ? Architecture and code map ? Security
• Official/Community publishing design ? Roadmap
• Example CSV

Development checks (Node 22+ is only needed for the client regression harness):

python -m unittest discover -s tests -q
node tests/test_ui.cjs
python scripts/check_frontend.py
python scripts/check_release.py
python scripts/build_release.py

License

Copyright (c) 2026 RayRayElite and contributors. Source code is licensed AGPL-3.0-only; see LICENSE. Modified network-served versions must meet AGPL source-availability requirements. Provider data, names and third-party artwork have their own rights and are not relicensed by this project's code license.