PlexInformationBot

A Discord bot that shows what your Plex server is doing: the number of people watching in the bot's status, a status message with the current streams, an optional Radarr/Sonarr download queue message, and a role message people can react to. The bot is set up on a small web page it serves itself, so no environment variables are needed on a new install.

Quick start with Docker

docker run -d --name plexinfobot -p 4546:4546 -v "$(pwd)/config:/config" redzguilt/plexinfobot

Instead of the config folder you can use a named volume, for example -v plexinfobot-config:/config.

1. Open http://localhost:4546.

2. Sign in with admin / admin. You are asked to choose your own username and password straight away (at least 8 characters).

3. Fill in the Discord bot token and your Plex address, port and token. Tautulli, Radarr and Sonarr are optional.

4. Click Test in each section to check the connection.

5. Click Save. The bot starts (or restarts) with the new settings.

6. In Discord, run these in the channels you want:

   • /setupreact (pick the role to hand out) posts the role message.
   • /setupstatus posts the status message.
   • /setupcontentqueue posts the download queue message.

   These commands are admin-only.

Invite link, health dots and setup checklist

• Invite link: once a Discord bot token is saved, the Discord section shows "Invite the bot to your server". It opens Discord's authorisation page in a new tab with the permissions the bot needs (view channels, send messages, embed links, read history, add reactions, manage roles). The link is built from the token on the server without contacting Discord, and the token is never sent to the page. It is hidden when there is no usable token.
• Health dots: each section heading has a dot and a label: green "Connected", red "Unreachable", grey "Not set" or "Checking". Hover for the message. The page asks every 30 seconds while it is visible and you are signed in. The server runs the same checks as the Test buttons in the background, at most once a minute (and not sooner than 10 seconds after the last one when you save), and answers from the last result, so a slow service never slows the page. Discord's dot reflects the running bot; it does not log in a second time.
• Setup checklist: a card under the status box lists the first-run steps: bot token saved and running, Plex connected, slash commands registered, status message set, and the optional download queue and role messages, each with a one-line hint when not done. It disappears once the first four are done, or when you press Hide (remembered in your browser). "Show setup checklist" in the status box brings it back.

Changing the port

The setup page listens on 4546. To use another port, set WEB_PORT and map the same port:

docker run -d --name plexinfobot -e WEB_PORT=8080 -p 8080:8080 -v "$(pwd)/config:/config" redzguilt/plexinfobot

If the port is already taken, the bot still starts and keeps running when a Discord token is configured, but without the setup page (check the log for the bind error). With no token configured it exits with an error, since there would be nothing to run.

WEB_PORT is the only environment setting a new install needs. CONFIG_DIR (default /config) can move the config folder but you normally do not need it.

Unraid

Add a port mapping for container port 4546 and a path mapping from /config to a folder in your appdata share. The container starts as root only to prepare /config, then runs the bot as PUID:PGID (default 99:100, Unraid's nobody:users). Set the PUID and PGID environment variables to use another user; PUID=0 keeps running as root. On every start the contents of /config are chowned to that user, so files written by older (root) versions stay readable. If /config is read-only the chown is skipped with a warning.

Using messages you already have

Normally /setupstatus, /setupcontentqueue and /setupreact post a new message and save its IDs. If the bot already wrote a status, queue or role message and you want to keep using it, turn on Advanced mode, click the Status, Queue or Role message item at the top of the setup page to open a popup for that message, and fill in its IDs instead.

• In Discord turn on Developer Mode (User Settings, Advanced), right-click the message and choose Copy Message Link. Paste it into "Paste message link" in the matching group and the channel and message IDs are filled in for you. The role ID is separate: in Server Settings, Roles, right-click the role and choose Copy Role ID.
• The IDs are the long numbers Discord shows (17 to 20 digits). Anything else is refused when you save.
• Click Test in a group to check, with the running bot, that the channel and message exist and that the bot wrote the message (a bot can only edit its own messages). For the role message it also checks that the role can be handed out. Test needs the bot to be connected, so save the Discord token first.
• Only fields you change are saved, so a message set up by a slash command in the meantime is not overwritten by an older copy of the page. Emptying a field clears it.

Advanced mode

The setup page starts in basic mode: only the Discord token and Plex sections are shown, and the status line at the top (Status, Queue and Role message) is plain text. Tick "Advanced mode" in the header to also show Tautulli, Radarr, Sonarr, and to make the status line items links. Clicking one opens a popup with just that group (status, queue or role); the popup has its own Save button, and Done, the X, Escape or a click outside closes it. The choice is remembered in your browser. Switching modes never clears anything: fields that are hidden keep their saved values.

Logs

The bot writes its log to bot.log in the logs folder inside the config folder (/config/logs/bot.log by default, readable only by its owner). It rolls over at 2 MB and keeps 3 older files (bot.1.log to bot.3.log). Console output is unchanged.

Click "Logs" in the setup page header to watch the latest bot.log live. It refreshes every 2 seconds while open, scrolls with new lines unless you scroll up, and has a Pause button. Only the last 1000 lines are kept in the page.

Secrets are redacted before the log leaves the server: Plex tokens, token and API key values in URLs or headers, and every token or key saved in your settings are replaced with ****. Redaction also covers the written file for token and API key parameters, but the file is still your private data, so keep the config folder private.

Security notes

• Open the page right after starting the container and change the default login. Until you do, anyone who can reach the port can sign in as admin / admin.
• Never forward the port to the internet. If you must expose it, put it behind your own reverse proxy with HTTPS.
• Behind a reverse proxy that rewrites the Host header, signing in can return 403. The page checks that the request's Origin matches its Host, so pass the original Host through. For nginx: proxy_set_header Host $host;
• To reset the login to admin / admin, delete credentials.json in the config folder.
• The Discord bot token, Plex token and API keys are saved in settings.json in the config folder. Keep that folder private. The page only shows saved secrets as ********; the eye icon next to a saved secret reveals it to a signed-in user, on click only.

Existing installs

The old environment variables still work as a fallback. Anything saved on the setup page wins, field by field. An empty field on the page counts as unset, so the environment variable is used.

Building from source

Requires Java 17.

./gradlew fatJar
java -jar build/libs/PlexInformationBot.jar

To build the image yourself: docker build -t plexinfobot .