How to do everything with Panda.

Panda doesn't have a one-click invite URL — you add it from inside Echoed, the way you'd add any other bot, then come here to configure it. These five steps get you live with sensible defaults.

1. In Echoed, open the server you want Panda in. Click the server name at the top of the channel list, then Server Settings.
2. Click the Bots tab. You'll see this tab only if you have Manage Server permission — if you don't, ask your server owner to add Panda for you.
3. Find Panda in the list of available bots and click Add. Echoed shows you what permissions Panda is requesting — review them and confirm. Panda appears in your member list immediately.
4. Come back to this site (memebot.echoed.gg) and click Login with Echoed in the top right.
5. Pick your server from the sidebar. The first time you land, Panda's already running with sensible defaults — type !ping in any text channel to confirm it's responding.

That's it. Everything else on this page is fine-tuning.

You add Panda from inside Echoed — there's no separate invite URL or OAuth popup. The full flow is below. You'll need Manage Server permission on the server. If you don't, ask the owner to add it for you.

1. Open Echoed and go to the server you want Panda in.
2. Click the server name at the very top of the channel list — a menu drops down.
3. Pick Server Settings. The settings panel opens.
4. In the settings sidebar, click the Bots tab. (If you don't see this tab, you don't have Manage Server permission — see the heads-up below.)
5. Browse the available bots, find Panda, and click Add.
6. Echoed shows you what permissions Panda needs. Read them — you can untick anything you don't want, but some features will stop working without their permissions (e.g. untick Manage Roles and reaction roles + level rewards stop working).
7. Confirm. Panda joins your server within a second or two and shows up in the member list.
8. To configure Panda, come to memebot.echoed.gg and click Login with Echoed. Your server appears in the dashboard sidebar.

Every feature has a dedicated page. The sidebar on the left lists them all; the main panel shows whichever one you've clicked. Whatever changes you make are not saved until you press Save at the bottom of the page — you'll see a yellow bar appear once you've edited anything.

If you switch pages with unsaved changes, the dashboard warns you before discarding them. Press Discard to throw them away or Stay to go back.

Most features can be restricted to specific roles or channels. The pattern is the same everywhere in the dashboard:

1. Allowed roles — only members with at least one of these roles can use the feature. Leave empty to allow everyone.
2. Ignored roles — members with any of these roles are blocked from the feature, even if they're in Allowed roles. Useful for muting bots from earning XP.
3. Allowed channels — the feature only fires in these channels. Leave empty to allow everywhere.
4. Ignored channels — the feature is suppressed in these channels. Useful for excluding spam channels from XP or auto-mod.

Ignored always wins over Allowed. So if you put @bots in Ignored roles, members with that role never trigger the feature, even if they have other roles in Allowed.

Panda plays YouTube and SoundCloud audio in any voice channel. It joins the channel you're in when you run !play, so you don't have to specify one.

DJ role: if you set a DJ role on the dashboard, only members with that role can skip, pause, stop, or change the volume. Everyone can still queue songs and use !rank-style lookup commands.

1. Open the dashboard → Music.
2. DJ role — pick a role (or leave empty to let anyone control playback).
3. Default volume — 50 is a reasonable starting point; persists per server.
4. Click Save.
5. Join a voice channel and type !play lofi study mix. Panda joins, queues the result, and starts playing.

Members earn XP for each message they send. As they level up you can grant them roles — that's how you build a "regulars" or "veteran" tier without manually assigning roles every time someone hangs around.

Anti-spam decay is built in: once a member sends a message they have to wait 60 seconds before another message earns XP. Bots and ignored channels never earn XP.

1. Open the dashboard → Levels and toggle Enabled on.
2. XP per message — defaults to 15-25 (random in that range). Lower numbers slow progression; higher speeds it up.
3. Ignored channels — add channels where messages shouldn't earn XP (spam, bot-commands, etc).
4. Role rewards — click Add reward. Pick a level (e.g. 5) and a role (e.g. @member). When someone hits level 5 they automatically get @member.
5. Stack roles — if on, members keep all reward roles they've earned. If off, they only have the highest one.
6. Click Save.

When a new member joins, Panda can post a greeting in any channel and optionally assign them a starter role.

1. Open the dashboard → Welcome and toggle Enabled on.
2. Channel — pick where the welcome message should appear (usually #general or a dedicated #welcome).
3. Message — write the greeting. Use {user} to mention the new member, {server} for the server name, and {count} for total member count.
4. Auto-role (optional) — pick a role to assign new members automatically. Useful for verification gates.
5. Click Save, then Test to see what the message looks like.

Members assign themselves roles by reacting to a message with a specific emoji. Common uses: pick your colour, opt into a ping role, choose a pronoun.

1. Post a message in your server explaining the choices (e.g. "React with 🔴 for red, 🔵 for blue").
2. Right-click that message and copy its message ID. (If your settings don't show "Copy ID", turn on Developer Mode in Echoed's user settings.)
3. Open the dashboard → Reaction Roles and click New panel.
4. Paste the message ID, pick the channel, then add each emoji + role pair.
5. Mode: Toggle — react to add the role, react again to remove it. Verify — react once to add, can't unassign. Pick one — only one role from the panel at a time.
6. Click Save. Panda adds its reactions to the message.

Run reaction-based giveaways with auto-pick winners, configurable duration, and entry requirements (minimum account age, required role, etc.).

1. Open the dashboard → Giveaways and click New giveaway.
2. Channel — where the giveaway message will be posted.
3. Prize — short text describing what's being given away.
4. Duration — when the giveaway ends. Format: 3d for 3 days, 12h for 12 hours.
5. Winners — how many people win.
6. Requirements (optional) — required role, minimum account age in days, etc.
7. Click Start. Panda posts the giveaway and members enter by reacting with 🎉.

Eight filters that watch every message and act when something crosses your threshold. Each one is independent — you can turn any on or off without affecting the others.

The filters: spam (rapid messages), caps (high uppercase ratio), links (URLs), invites (Discord/Echoed invites), mass-mention, emoji-spam, zalgo (corrupted text), bad-words (your custom list).

1. Open the dashboard → Auto-mod.
2. For each filter you want active, toggle it on and set the threshold:
   • Spam: N messages within M seconds
   • Caps: percentage of uppercase letters in messages over X characters long
   • Bad-words: add your own list one per line; supports partial matches
3. Action per filter: delete, warn, timeout, or kick. Choose what happens when the filter fires.
4. Exempt roles — moderators and trusted members. They never trigger filters.
5. Click Save.

Panda adds reactions automatically to messages matching patterns you specify. Useful for: marking announcements with a starter set of reactions, adding 👀 to every message in #suggestions, etc.

1. Open the dashboard → Auto-react and click New rule.
2. Match — choose any message in channel (always reacts) or contains text (matches a keyword).
3. Channel — restrict to one channel, or leave empty to apply server-wide.
4. Reactions — pick one or more emoji. Custom server emoji work.
5. Click Save.

Define your own !commands that send a fixed response. Common uses: server rules, FAQ replies, role-info cheatsheets, recurring jokes.

1. Open the dashboard → Custom Commands and click New command.
2. Trigger — the command name without the !. E.g. rules means members run !rules.
3. Response — what Panda replies. Use {user}, {server}, {args} placeholders.
4. Restrictions (optional) — limit to specific channels or roles.
5. Cooldown (optional) — minimum seconds between uses, per member.
6. Click Save. The command works immediately.

Post messages on a schedule — daily reminders, weekly event announcements, monthly check-ins.

1. Open the dashboard → Schedules and click New schedule.
2. Channel — where the message gets posted.
3. Message — what to send. Mentions and links work normally.
4. Frequency — once, daily, weekly, or custom (cron-style: 0 9 * * 1 = 9am every Monday).
5. Timezone — defaults to UTC. Pick your server's primary timezone for sane "9am" expectations.
6. Click Save.

Get a DM when specific words show up in your server's chat. Useful for catching mentions of your name, your project, or a topic you care about — without scrolling everywhere.

1. Open the dashboard → Keywords and click Add keyword.
2. Keyword — the word or phrase to watch for. Case-insensitive. Whole-word matching by default.
3. Notify — yourself, or a list of members.
4. Restrict to channels (optional) — only watch in specific channels.
5. Click Save. You'll get a DM the next time the word appears.

Manual mod actions with full audit logging. Each action gets a case number and shows up in your audit channel with the actor, target, reason, and any escalation context (e.g. "3rd warning in 30 days").

1. Open the dashboard → Moderation.
2. Audit channel — pick a private channel only mods can read. Every mod action gets logged here.
3. DM target on action — if on, Panda DMs the kicked/banned member with the reason. If off, they're silently removed.
4. Escalation thresholds (optional) — e.g. 3 warnings in 30 days = auto-timeout. Configurable per action.
5. Click Save.

Watches the rate at which new members are joining and the average age of their accounts. When something looks off, Panda enters lockdown: invites pause, verification level climbs, new joiners go to a holding queue you review when you're back.

1. Open the dashboard → General → Anti-raid.
2. Join rate threshold — N joins within M seconds triggers lockdown. Default 12 / 60 catches most automated raids without false-positiving viral growth.
3. Account age threshold — accounts younger than this go straight to the queue. Default 24 hours.
4. Lockdown notification channel — where Panda pings you when lockdown engages.
5. Auto-lift — minutes after the last suspicious join with no further activity, lockdown lifts automatically. Default 30. Set to 0 to require manual lift.
6. Click Save.

Live-updating voice channel names that show server stats — member count, online count, custom values you push from outside.

1. Create a voice channel that nobody will join (e.g. 📊 Members: —). Members listening to the channel name = members not connecting to it.
2. Open the dashboard → Stat Counters and click New counter.
3. Channel — pick the voice channel from step 1.
4. Stat — total members, online members, member count of a specific role, or custom (push values via the API).
5. Format — e.g. 📊 Members: {value}.
6. Click Save. The channel renames within a minute.

Watch Twitch streamers, YouTube channels, and Reddit feeds. Panda posts in your channel when they go live or upload something new.

1. Open the dashboard → Notifications and click New watcher.
2. Service — Twitch, YouTube, or Reddit.
3. Account / channel / subreddit — paste the username, channel handle, or subreddit name.
4. Notify in — pick the channel where the alert posts.
5. Mention (optional) — role to ping with each alert. Many servers create a @stream-pings role members opt into via reaction roles.
6. Custom message (optional) — override the default text. Use {title}, {url}, {author}.
7. Click Save.

Most "Panda isn't responding" reports come down to one of three causes. Walk through these in order before opening a bug report.

Still stuck? Open an issue on GitHub with: what you tried, what you expected, what happened instead, and your server ID (visible in the dashboard URL). We read every one.