LatestArr collects everything added recently across your media servers and libraries and emails it to the people you choose, on a schedule you set. Design the email with options and a live preview, or hand-write it in MJML. It runs as a single Docker container with SQLite, on Intel or ARM, backs itself up nightly, and supports multiple users with roles, SSO, failure alerts, and imports from your server’s users or a spreadsheet.
Sources: Plex, Tautulli, Jellyfin, Emby, BookLore, BookOrbit, Grimmory, Audiobookshelf and RomM.
Highlights:
- Designs: an options-based design editor with a live preview, light and dark mode, logos, and Where to watch buttons, plus a code mode in MJML and Handlebars.
- Scheduling that looks after itself: daily, weekly or monthly sends, catch-up after downtime, skipping empty issues, and "send to the rest" after a partial send.
- Recipients and groups: import from your media server's users, or from a pasted list, CSV or Excel file, with a review before anything changes and an Undo.
- Team-ready: admin, editor and viewer roles, SSO with any OIDC provider, and a recovery path when someone's locked out.
- Reliable: automatic nightly and pre-upgrade backups, failure alerts by email, Discord, Slack, ntfy, Apprise or webhook, and a Logs page.
Background
Tautulli's newsletter was the closest thing to what I wanted, but it only speaks to Plex and gives limited control over layout, scheduling and branding. LatestArr is Plex-aware but not Plex-only: it covers the rest of a self-hosted library too, and lets you design the email yourself.
Why this project
It's a way to give something useful back to the self-hosted and homelab community. Security and accessibility were part of the plan from the start, not added later:
- Security: credentials encrypted at rest, server-side sessions, OIDC/SSO, rate limiting, CSRF protection, and input validation on every route.
- Accessibility: automated axe-core checks on every page, and keyboard-only testing of the admin UI.
How it's built
AI is LatestArr's lead developer. The code, tests, CI and docs are written by Claude Code, Anthropic's AI coding agent. I direct the work, review every change before it merges, and test releases on my own media servers.
What keeps it honest:
- Automated checks on every change: lint, typecheck, build and tests on Node 24 and 26, plus a Docker image built and smoke-tested for Intel and ARM.
- Review sized to the risk: changes to limits, fetching, sending, links, sources or sign-in get a second review pass and a real send before release.
- Real users: feedback from people running it has shaped recent releases.
More on this on latestarr.app.
Where it's at
LatestArr is pre-1.0: usable today, with frequent releases and the occasional rough edge. The main gaps before 1.0:
- Unsubscribe links (planned for 0.13.0). Today, recipients are turned off by hand.
- More real-world testing: Plex (direct), BookLore, Grimmory and Audiobookshelf need testers, and Jellyfin and Emby haven't been confirmed on a real server yet.
Contribute
LatestArr is open source under GPL-3.0. New sources are self-contained adapters, and testing, code review, issues, pull requests and adapter requests are all welcome. See Contributing.