Argus — All-Seeing Eye for ARR Stack Gaps

Argus helps find and fetch missing content for Sonarr, Radarr, and similar ARR tools by searching YouTube for missing episodes/movies and ingesting them into your collection.

Key Features

• Search and find missing content on YouTube.
• Download content via yt-dlp.
• Transcode content using FFmpeg to configurable specs.
• Produce artifacts for final import by Sonarr/Radarr.

Use docs/issues/ to draft issues that are automatically converted into GitHub issues via .github/workflows/docs-issues.yml.

Getting Started

This repo contains an agents/ directory that defines a simple agent architecture used for project tasks. Use the agents/manager.py script to run and orchestrate tasks locally.

This project uses the copilot-instructions.md to define how Copilot and contributors should interact with the repo.

Quick run (development)

Requirements: Python 3.10+, yt-dlp and ffmpeg installed.

Run manager with a sample task:

python3 agents/manager.py --example

Docker image

Argus can be built and run as a Docker container. The repository includes a Dockerfile and a GitHub Actions workflow to build and publish images to GHCR.

Build locally:

docker build -t argus:local .
docker run --rm -p 8000:8000 argus:local

Run the manager with the example task (default):

docker run --rm -p 8000:8000 argus:local
# Or run the manager with the UI running
docker run --rm -p 8000:8000 argus:local --example --webui

ARR integration and Web UI

Use the Web UI to configure your Sonarr or Radarr instance: open http://localhost:8000 and navigate to the ARR Settings panel. Provide the API URL and API key to enable the ARR integration. You can use the "Test ARR" button to validate connectivity, and the "Search Missing (ARR)" button to generate search tasks for missing items.

Override the entrypoint to run a different command:

docker run --rm argus:local --example

Pull from GitHub Container Registry
---
Images are published to GitHub Container Registry under `ghcr.io/<owner>/argus` by the `Build and Publish Docker image` workflow. Example usage:

```bash
docker pull ghcr.io/<owner>/argus:latest
docker run --rm ghcr.io/<owner>/argus:latest

Contributions
---
See `CONTRIBUTING.md` for development conventions, PR workflow, and agent contract guidelines.