ShowFlow

Automated media library manager. Track shows, define quality preferences, and let ShowFlow find, download, and upgrade episodes — similar to Sonarr/Radarr but self-contained in a single Docker image.

Status

Automatic episode grabbing and blackhole importing are functional. Quality engine with custom formats, upgrade logic, and profile management is in place. The release/supervisor pipeline (Docker deployment with over-the-air updates) is pending refinement and testing.

Quick Start

docker run -d \
  --name showflow \
  -p 3000:3000 \
  -v /path/to/data:/data \
  -v /path/to/media:/media \
  ghcr.io/PylotLight/showflow:latest

Open http://localhost:3000 — the wizard walks through basic setup (media paths, indexer, API keys).

Configuration

Configured via environment variables or the settings UI:

Variable	Purpose
`SHOWFLOW_DATA_DIR`	Database & state storage (default: `/data`)
`SHOWFLOW_DEBUG`	Enable debug logging (`true` / unset)
`TMDB_API_KEY`	Metadata & poster images
`TVDB_API_KEY`	TV metadata (TheTVDB)
`TVDB_PIN`	TVDB API PIN (if required)
`http_proxy` / `https_proxy` / `all_proxy` / `no_proxy`	HTTP proxy configuration

Architecture

Backend: Bun + SQLite — quality engine, show resolution, indexer integration
Frontend: React (single HTML import) — show management, calendar, settings
Supervisor: Standalone binary — port proxy, graceful handoff, over-the-air updates

See docs/arch.md for the full architecture and API reference.

Caveats

TV shows only — movie support is planned for a future release.
Single-user — no multi-user or permission system.
Beta quality — the release/update pipeline is still hardening; manual image pulls are the stable path.

PylotLight/ShowFlow

README