YAWBDB

YAWBDB — Yet Another Web Based Discord Bot

YAWBDB is an all-in-one Discord bot with a full web panel — configure everything from your browser, not from slash commands.

Everything YAWBDB does

Thirty-seven features, each one independent — enable only what you need. The headliners are detailed further down; every row links to its full documentation.

Feature	What it does
🏆 Leveling & XP	Text, voice and reaction XP, streaks, prestige, decay, image rank cards, public web leaderboard, a visual reward-graph editor, and XP import from MEE6 or a JSON export.
🛡️ Moderation	Ban, mute, kick, warn, softban from the panel or Discord, with full history, auto-expiry and a built-in appeal system.
🤖 Automod	Nine rule types — bad words, spam, mass mentions, links, invites, caps, new accounts, anti-scam and anti-poll — each with its own action.
🚧 Anti-Raid	Detects mass-join bursts and answers with kick, ban or full lockdown, with an always-on minimum account age at the door.
🧨 Anti-Nuke	Watches your audit log for a rogue admin — mass bans, kicks, channel or role deletions — and neutralises the actor automatically.
🚪 Invite Guard	Decide which roles and members may create an invite to your server; every other invite is revoked as it appears. Discord's own widget invites are recognised and left alone.
🍯 Honeypot	Trap channels, trap roles and trap invites that sanction spammers and self-bots on sight, no moderator needed. Several traps in one window ask Anti-Raid for a lockdown. A staff handout is never a catch, and an invite it cannot attribute sanctions nobody.
🚨 Discord Reports	/report opens a private channel for your team with action buttons, fully synced with the panel.
🎫 Discord Tickets	/ticket support channels with categories, staff roles, rate limits, internal notes and .md/.html export.
🤖 AI Responses	Bring your own OpenAI-compatible key, or your own endpoint. The model is picked from the list your endpoint publishes rather than typed from memory. Persistent server memory you can inspect and wipe, thread awareness, and the judgement to stay silent.
🎙️ Voice & Audio	Record voice channels on demand, transcribe audio with the model of your choice, and generate AI meeting summaries.
🔊 Temporary Voice	Join-to-Create rooms members own and manage with /voice, deleted automatically when empty.
💬 Panel Chat	Direct messages, groups and real-time sync inside the web panel — markdown, reactions, replies, GIFs, URL embeds.
📋 Auto Responder	Custom responses to words, phrases or regex, mapped to channels with a drag-and-drop interface.
⌨️ Custom Commands	Your own slash commands, no coding — static replies, or a visual node graph for real behaviour.
✨ Graph Assistant	Describe the automation you want; it edits your reward graph and proposes the changes. You review them one by one and apply — it never changes anything on its own.
🛠️ Config Assistant	Ask what a setting does, what it currently holds, and what to change to get a different behaviour. It reads your configuration, explains it, and can propose a change as a before/after list — nothing is applied until you press Apply.
🎨 Embed Builder	Compose real Discord embeds with a live preview, save them as templates, reuse them across the panel.
📝 Form Builder	Up to five typed questions in a native Discord pop-up — short, long, single choice, multiple choice — opened with /form open and reviewed in the panel.
🎭 Reaction Roles	Self-assign roles by reaction or button. Four modes: add-only, toggle, unique and verify.
💡 Suggestions	/suggest with 👍/👎 voting, staff statuses that sync back to the embed, optional threads and anonymous mode.
✅ Join Verification	A button — and optionally a captcha — before your server unlocks. Optional auto-kick for accounts that never verify.
⏰ Scheduled Messages	One-off or recurring posts that survive restarts, with a rich embed instead of plain text or alongside it.
🔔 Social Alerts	YouTube, Twitch, Reddit, Bluesky and any RSS feed, announced in the channel of your choice.
⭐ Starboard	Star a message, and past your threshold the bot reposts it with a live count and a jump link.
🎂 Birthdays	Saved once with /birthday set and remembered across servers. Announced on the day, with an optional temporary role.
🎉 Giveaways	One-click entry, weighted bonus entries, role and level gating, drawn panel-side so they always end on time.
📋 Server Logs	Ten independent categories, each with its own channel, stored for up to 365 days with a searchable viewer — channel permission overwrites and a change of server owner included.
📜 Audit Log	Every action taken in the panel, with a before/after — per server, per team, and a full admin view.
🪝 Outgoing Webhooks	Your team's activity pushed to a URL you own, signed with HMAC-SHA256, retried, deduplicated — no polling.
📊 Counters	Voice channels that rename themselves with live stats — members, bots, online, boosts, bans and more.
🏅 Statroles	Roles granted on lifetime totals — messages sent, voice time, games won, days on the server — independent of the XP curve.
🎮 Games	The daily Wordle, playable as a command or in an embedded Discord Activity — plus Hangman, Rock Paper Scissors, Trivia, Bingo, an 8-ball and a love calculator.
🗳️ Anonymous Voting	Server-wide votes from a link, Discord login required, results fully anonymous.
📊 Server Insights	A full analytics dashboard per server, with KPIs, trends, an online-members curve and a 7-day forecast.
👋 Welcome & Auto Role	A generated welcome image with the member's avatar, plus an automatic join role.
📜 Rules Channel	Keep a rules channel in sync from the panel — edit once, the bot maintains the message.

Plus random bot interactions, audio-to-text in configured channels, and an onboarding guide for new server owners.

---

The headliners

🏆 Leveling & XP A fully customizable XP system with text, reaction, and voice activity tracking. Daily streaks, prestige resets, decay for inactive members, and a leaderboard.

Every night, the member who earned the most XP is crowned with a Daily Top Member role — and anyone can check who's leading with /top.

/rank and /leaderboard come with a sleek generated image card — avatar, level, rank, an XP progress bar, prestige, title and streak — fully customizable with your own accent color and background, or one of the ready-made themes (Halloween, Christmas, Valentine, Neon, Summer, Winter) with a live preview in the panel.

You can even publish a public web leaderboard anyone can open in a browser (no Discord login), shareable and search-indexable.

The reward system is built on a visual node graph — chain triggers, conditions, and actions to build any automation you want: assign roles at level 10, send a DM at level 50, grant a title at prestige, celebrate the day's top member, boost XP during timed events, and more.

Every card in that editor now carries an (i) button telling you what it reads, which variables you can type into its message, and — the useful part — the one empty field that would make it do nothing at all. A card missing its role or its message is skipped in silence when the graph runs, so the editor points it out before your members do.

Already running another leveling bot? Import your members' XP instead of resetting everyone to zero — straight from MEE6, or from a JSON export of any other bot. You get a preview before anything is written, a choice about what happens to members who already have XP here, and their role rewards can come across too. Because two bots rarely share the same level formula, the preview measures the difference on your real data and lets you adopt the source's curve so nobody loses a level. The whole move is written up in Switching from MEE6 to YAWBDB.

🛡️ Moderation Ban, mute, kick, warn, and softban directly from the panel or with slash commands.

Full moderation history, automatic sanction expiry, log channel embeds, and a built-in appeal system — sanctioned users receive a personal appeal link by DM, your staff gets an email the moment an appeal is submitted, and your team reviews everything from the backoffice.

Set an appeal window and optionally auto-close appeals when it expires — with your own closing message — while moderators can always reopen a closed appeal.

🤖 Automod, Anti-Raid, Anti-Nuke & Invite Guard Automatic moderation, 24/7. Nine rule types — bad words (with regex), spam, mass mentions, link & invite filters, caps, new accounts, anti-scam (a built-in curated list of fake-Nitro / gift / crypto-drainer scams and lookalike domains, extendable with your own patterns) and anti-poll (block native Discord polls outside allowed channels) — each with its own action: delete, warn, mute, kick, ban, or flag for review.

Every action is logged and lands in the member's moderation history, same as manual sanctions.

Anti-Raid detects mass-join bursts and responds with kick, ban, or full lockdown, with an always-on minimum account age filter at the door.

Anti-Nuke covers the threat Anti-Raid can't catch, because it comes from someone with power: it watches your audit log and counts destructive actions per actor — mass bans, mass kicks, mass channel or role deletions — within a sliding window. Cross the threshold and the bot strips their roles, kicks, or bans them, then alerts your team. The bot, the owner, and your whitelisted roles are always exempt.

Invite Guard takes back a decision Discord doesn't let you make: Create Invite is a single permission, on or off, and anyone who has it can invite anyone, from any channel, forever. Here you list the roles — and the individual members — allowed to create an invite, and every other invite is revoked the second it appears, with the code, the channel and the creator written to your log channel.

Enabling it can't lock you out. The bot, the owner, your server managers and the applications you authorized keep their right by default, and you remove those exemptions one at a time when you're ready. The invites Discord makes on its own — the server widget, Server Discovery — have no author and come back the moment somebody loads the widget, so they're recognised and left alone rather than fought with; and if five invites are revoked in a single minute the guard stands down and tells you where to look instead of trading deletions with whatever is recreating them.

🎫 Reports & Tickets Members report users with /report — the bot opens a private channel for your moderation team with action buttons (review, resolve, dismiss, sanction). Members open support tickets with /ticket — private channels, configurable categories and staff roles, rate-limited per user.

Both sync bidirectionally with the panel: staff replies appear as embeds in the channel, status changes reflect instantly on both sides. Tickets add internal staff notes and export as .md or .html.

🤖 AI-Powered Responses Connect your own OpenAI key (or any compatible endpoint — Ollama, LM Studio, OpenRouter) and let the bot respond intelligently when mentioned. The bot builds a persistent memory of your server's context across conversations.

It follows conversation threads — it sees which message a member is replying to — and knows when to stay silent: if a message doesn't call for a response, it can simply react or say nothing instead of butting in.

It won't talk to itself either — replying to its own messages is off by default, and you can toggle it per server.

Nothing is ever forgotten behind your back. No expiry, no automatic clean-up: a fact stays until the bot decides it no longer holds, or until you remove it yourself. How much of it reaches the AI is your call — Max memory facts injected, on the feature's settings page.

And the bot can now act on that judgement. Each fact reaches it with its own number, so when something stops being true it can drop that exact one — where before it could only be told it was allowed to forget, without ever being shown what to point at. It can only drop a fact it was actually shown in that exchange, so a number it made up touches nothing.

How it remembers is now yours to choose. Memory type, on the AI Responses settings page, is not a dropdown: click it and six ways of remembering open over the page, side by side, each with what it does, what it would change, and where to read more.

Journal is the second of them, and it is available now. A fact the bot decides is out of date is no longer erased — it is retired: it leaves the AI's context and stays on your memory page, struck through, with the date it stopped and the fact that replaced it. The bot gains the move to match, saying "this is no longer true, here is what replaces it" — and it is told the old wording survives, because a correction it believes is destructive is a correction it hesitates to make. Got it wrong? Restore puts the fact back in one click.

Turning it on moves nothing: everything you already have counts as still true and keeps the date it was created. Classic stays the default, and a server that never opens the field behaves exactly as it always has.

Searchable archive is the third, and it is available now too. Until now, everything past Max memory facts injected was invisible to the AI however true it was, and raising that limit made every message more expensive. Now the bot can simply ask — "search the memory for: Paul's birthday" — and it is handed the matching facts before it answers.

It costs nothing on the messages where it does not look anything up, and exactly one extra call when it does. The search is full text, run by the database over what you already have: no embeddings, no vectors, nothing to compute. And what the bot finds by searching, it can also correct.

Before you switch, the memory page tells you what it would cost in your own numbers: how many facts there are to index, how much text that is, and how long a real search over them just took.

Ranked memory is the fourth, and it is available now. Classic hands the model the last few facts modified — which quietly means a rule your server has lived by for a year loses its place to yesterday's small talk. Ranked scores every fact on how recent it is, how much it matters, and how much it has to do with what is being said, then hands over the best ones. It is a ranking and not a filter: a fact that misses the cut on one message is a candidate again on the next.

The three weights are yours to set, per server, and switching costs nothing — a fact nobody has scored yet ranks in the middle rather than last, so it works from the first message on the memory you already have.

Consolidated memory is the fifth, and it is available now too. When your server goes quiet, a background pass goes over its memory: it brings duplicates together — language and lang becoming one line instead of two — and flags the facts that contradict each other rather than picking a winner.

It proposes, and you decide. Every pair lands on your memory page with three answers: merge, merge the other way round, or keep both. Nothing is applied until you click, and a yes never erases — the fact that steps aside is retired, with Restore one click away.

It rewrites your server summary too, which is the one line the bot reads whole every time and the one that goes stale first. The page shows what your bot reads today next to what it would read instead, and again, nothing replaces it without your yes.

The pass runs on its own model, configured separately with its own endpoint and key — it is not racing a Discord reply, so it can be a stronger or a cheaper one. It works with no model at all, too: a merge then simply keeps the winning fact word for word.

Semantic memory is the sixth, and the last one to arrive. The five before it all come down to the same question underneath — do these two texts share words — and for "he speaks French" and "the server language is French" the answer is no. This one compares meaning instead: each fact becomes a direction, and two facts that say the same thing point the same way.

Two things change. The bot's search stops missing what was worded differently. And the facts it is handed on each message stop being chosen because they were found and start being chosen because they are about what is being said.

It is the strongest of the six and the only one with a bill attached, so the numbers are on the card before you switch: one call per fact written, one per message long enough to be worth ranking, and — for anything shorter than a few words — none at all. If no model is named, or the endpoint refuses, or the catch-up pass has not finished, the word search takes over on its own and the answer tells you which one ran.

And you can wipe it: Clear memory empties everything the bot knows about a server, or everything it knows about one member — for someone who left, or asked to be forgotten. It takes the same permission as configuring the server, and it lands in your audit log with who did it and how many entries went.

✨ Graph Assistant Reward graphs are powerful and fiddly. Tell the assistant what you want instead — "give the Veteran role at level 25, but only to boosters" — and it drafts the graph for you.

It edits your graph rather than replacing it. It sends only the changes — add this card, change that field, wire these two — so anything it does not name is left exactly as it was. When a model ignores that and answers with a whole graph anyway, the answer is still honoured, but it is labelled as such and the review window lists everything that would disappear.

It proposes, never applies. A review window lists every card added, every field changed with its old and new value, and everything removed, checked against the graph you have on screen at that moment. Apply from there, or discard. Before you even get that choice, the proposal is checked against the real card list: an invented card, or an action left without its role, is reported instead of offered — those are the mistakes that would otherwise save fine and then silently never run.

And when a proposal is turned down, it corrects itself. The refusal goes back to the assistant with the exact edits that caused it, and it gets two more attempts before the answer reaches you — each one visible in the thread as it happens, so you watch it converge instead of having to explain the same problem twice.

Keep a conversation per subject — reward graphs and custom commands each get their own, and each keeps its own history. Ask again re-opens your last question so you can adjust a word instead of retyping it.

The threads belong to your whole team, live. Everyone sees who asked what, the model's reasoning is one click away under each answer, and while it is working the composer is disabled for everybody, in every conversation, so two requests never fight over the same canvas.

It can look across every server your team owns — so "set this up like my other server" is a question it can actually answer — while your API keys and tokens stay masked from it, and it can only ever write a graph on the server you have open.

You decide what it is allowed to read. Two switches in the chat window, yours alone and remembered across servers and conversations: the inventory of your team's servers, and what the bot has noted about the server you have open. Turn one off and it is not sent at all — not sent and ignored, simply never built.

Bring your own model if you prefer: set it on your team settings page — your own key, or your own OpenAI-compatible endpoint, as long as it is reachable over HTTPS at a public address. List models asks that endpoint what it serves and offers the answer, so the model is picked rather than typed from memory.

🛠️ Config Assistant Your bot has a lot of settings. This one sits on your server's feature configuration page and answers questions about them: what a setting does, what it currently holds, and what to change to get the behaviour you are after.

It reads your configuration as it is right now — not the documentation, your server. So "why does the level-up message go to the wrong channel" is a question it can actually look at.

Ask it for a change and it proposes one. Review changes opens a before/after list — the feature, the setting, the value now and the value it suggests — and nothing is written until you press Apply. Discard turns it down, and so does simply asking again: the older proposal is marked as replaced and can no longer be applied.

The list is recomputed when you open it and again when you apply it, so a proposal from an hour ago can never quietly undo what a teammate changed since. Applying it needs the same permission the settings form needs, it lands in your audit log under your name, and if any part of it does not fit, the whole thing is refused rather than half-written.

It also knows what you have not turned on. Half the answers to "how do I do X" are "that feature is not enabled here", and it says so instead of walking you through settings you cannot reach yet.

Channels and roles come back by name rather than as raw Discord ids, and your API keys and tokens are masked from it exactly as they are everywhere else in the panel.

Conversations are kept per subject and shared with your team, so a question about tickets does not land in the middle of one about levels.

⌨️ Custom Commands & Embed Builder Create your own slash commands for your server, no coding required. Members run /yourcommand and the bot replies with a text message or a rich embed. Restrict a command to specific roles, add a per-user cooldown, and make the reply private. Commands are registered on your server automatically.

Need more than a static reply? Switch a command to graph mode and design its behavior in a visual node editor: chain conditions (roles, booster status, level, account age, time windows…), logic gates, random branches, delays and cooldowns into actions — reply with dynamic variables, assign or remove roles, grant XP, send DMs, post to a channel, and more.

Embeds are composed in the Embed Builder, with a live preview matching exactly what your members will see: a title with a clickable link, a markdown description, a colour, an author with an icon, an image, a thumbnail, a footer, a timestamp, and up to 25 fields you can reorder and lay out full-width or side by side. Save them as reusable templates, or post one to a channel straight away with Send now.

The same editor powers Scheduled Messages and birthday announcements — you learn it once.

Prefer the old way? Set a text prefix for your server and your commands answer to it too: !rules as well as /rules. The games and the leveling lookups come along. Moderation deliberately does not — a slash command carries the permissions Discord itself enforces, and a text message would hand everyone a way around the restrictions you set.

📝 Form Builder Build a form in the panel and members fill it in from a real Discord pop-up — not a chain of questions in a channel. Short answers, long answers, single choice and multiple choice, up to five questions.

Members open one with /form open, and the name completes itself as they type — no need to remember what you called it.

Every submission arrives as a clean recap in the staff channel you pick, and is stored in the panel where you accept or reject it with a note.

Guard a form behind a role, put it on a cooldown, or limit it to one answer per member — then close it in one click when you have collected enough.

It deliberately stays out of the way of its neighbours: no voting, that is Suggestions, and no reply thread, that is Tickets. A staff application is a form; "I have a problem" is still a ticket.

🎂 Birthdays Members save their birthday once with /birthday set, and the bot wishes them a happy one on the day, in the channel you pick.

The date is remembered across every server the bot is in, so a member who has already entered it elsewhere types nothing — they just choose which servers get to announce it.

You set the channel, the hour and the timezone, an optional role to ping, and an optional birthday role the bot hands out and takes back on its own after the duration you choose.

Nobody can farm the spotlight: a member is announced at most once a year per server, no matter how often they rewrite their date. You can also require a minimum time on the server before anyone is eligible, and lots of birthdays on the same day go out as one grouped message instead of a burst of pings.

Everyone keeps control of their own data — /birthday remove opts out of one server, /birthday forget erases it everywhere.

🎮 Games — Wordle & more A growing hub of community mini-games, each individually toggleable with its own settings.

First up: Wordle — one five-letter word a day, the same for everyone on every server, resetting at midnight UTC. Six tries, private ephemeral grid rendered as an image, spoiler-free emoji sharing, streaks (NYT rules), per-server or global play, server leaderboard, and Leveling XP on solve if you want it.

Guesses are checked against a large English dictionary, and an optional practice mode (/wordle practice) offers unlimited off-record games — random words, no stats or streaks at stake.

While someone plays, an optional live-progress embed fills up in the channel in real time — spoiler-free, just like the official Wordle. Every day, an optional recap posts yesterday's results with a generated image of everyone's grids side by side, and a reset reminder can announce the new word — optionally pinging yesterday's top 3 or all of yesterday's players so nobody breaks their streak.

Play it as a slash command — or in a real embedded window inside Discord (Discord Activity) with the full grid and keyboard.

Alongside it, a set of quick mini-games for the in-between moments: Hangman (guess the word letter by letter from a drop-down, with a category as a hint), Rock Paper Scissors against the bot or as a duel against another member, Trivia where the first correct answer out of four wins, plus a magic 8-ball and a love calculator.

And a full Bingo: members join with a button, each gets a private card, and the bot calls the numbers on a timer until someone completes a line. The draw runs server-side, so a restart never loses a game in progress.

Each game is switched on separately, can be restricted to specific channels, carries its own cooldown, and can grant XP to the winner.

Each one also decides what it sends to your Server Logs: nothing, wins only, or every single round — losses, draws and all. Per game, because a Wordle is once a day per member while a rock-paper-scissors fires every few seconds, and the two do not belong in your log at the same volume.

Every game feeds the Games tab of your server Insights — rounds played, win rate, unique players and a 30-day curve, per game.

Wins also count towards a role. The new games won Statrole metric adds up mini-game wins and Wordle solves together, so "Game Champion at 50 wins" is one rule, not one per game — and the role lands the moment the win happens, not on the member's next message.

📊 Server Insights A full analytics dashboard per server — tabs covering activity trends, leveling stats, moderation history, report & ticket pipelines, member engagement, and games.

Charts filterable by 7d / 30d / 90d, KPIs (messages, voice minutes, commands, active members, growth, prestige members, total XP), an online-members trend (hourly snapshots → daily peak/average, plus a live count), and a 7-day linear regression forecast for message activity.

Use /trends in Discord to render a 30-day chart as an image directly in your channels.

🗳️ Anonymous Voting Create server-wide votes accessible directly from a link — Discord login required, fully anonymous results, with proposal submission and approval by admins.

---

📜 Audit Log Every action taken in the panel is written down: who did it, when, and on what. It is the trail for your dashboard, not for Discord — Server Logs covers the second.

Entries carry a before / after table, so you read the fields that actually moved and what they moved from and to, rather than "settings saved". Nothing is listed unless it really changed, lists are compared as a whole, and a token or a secret is recorded as rotated — never as its value.

The same trail is readable at three scopes — one server, a whole team, or the platform — with the same filters everywhere: search, category, exact action, person, date range. An action by a platform admin on your server appears in your own logs too, tagged as such, so backoffice intervention is never invisible to you.

It now covers what it used to miss, which was most of the panel: accepting or rejecting an application and deleting its answers, sanctioning a member, rewriting why, clearing someone's history, closing a report or a ticket, ruling on an appeal — and every server feature, from automod rules and custom commands to giveaways, scheduled messages, birthdays and the leveling reward graph. Clearing a member's history used to leave nothing behind at all on a server with no Discord log channel.

Decisions taken about a person sit in their own moderation category rather than among settings changes — adjusting someone's XP, resetting them, or ruling on their suggestion belongs with sanctions, not with a colour picker.

🪝 And you can have it pushed to your own server. Point your team's trail at a URL you own and every action you subscribe to arrives as a JSON POST the moment it happens — no polling, no export. Each delivery is signed with HMAC-SHA256 over a timestamp and the raw body, so you can prove it came from us and refuse a replay; it carries a delivery id that stays the same across retries, so a retry is never counted twice; and a brief outage on your side is retried four times over twelve minutes rather than lost.

A webhook carries exactly what the page shows and nothing more — same rows, same fields, minus the IP address, which never leaves the panel. Pick the categories you want, keep up to three endpoints, and read the last twenty attempts on screen when something stops arriving. Sign-in events are deliberately not on the list. 👉 https://yawbdb.alama.host/wiki/core-features/outgoing-webhooks

👥 Team roles you compose yourself Your team is not stuck with three fixed roles. Name your own, describe them, and tick exactly what each one may do — view servers, configure features, turn them on and off, moderate, remove a server, delete appeal records, use the graph assistant, or manage the roles themselves.

Nothing changes for the roles you already had: everyone stays on the role they were on, and Administrator, Editor and Moderator keep doing what they did. From there you can edit them, or add the ones your team actually needs.

Handing out privilege has limits, and they are enforced rather than suggested: you can never grant a permission you do not hold yourself, nobody edits the role they are standing on, and a role somebody still holds cannot be deleted. Every change lands in your audit log with a before and after. 👉 https://yawbdb.alama.host/wiki/getting-started/team-roles-and-permissions

📦 Your data, on demand Anyone with an account can ask for a copy of everything YAWBDB holds about them, straight from their profile — no email to write, no form to fill in. It covers both halves of an identity: the panel account, and, once Discord is connected, levels, birthdays, games and moderation history across every server.

Where a record also names somebody else — the moderator behind a sanction, the member behind a report — their identity is blanked out, and the file lists which fields were removed. Your access to your own data does not extend to theirs, and it works the same way in both directions.

Exports are built one at a time, so the page tells you roughly how long it will take before you ask, then your position while you wait. An email arrives when the file is ready, so there is no page to keep open. It stays downloadable for seven days, then it is deleted. 👉 https://yawbdb.alama.host/wiki/user-features/download-your-data

---

🌍 In your language, down to the email The panel speaks 13 languages, and so do the emails it sends you — verification, password reset, ticket, report, appeal, data export. Each one is written in the language your account chose, not the one the server happens to run in.

---

🔗 Features that work together YAWBDB features connect: Leveling reward-graph triggers fire on giveaway wins, approved suggestions, solved Wordles (with conditions on tries and streak) and members' birthdays — hand out bonus XP or a personal multiplier on their day; sanctioned members can be excluded from draws; engagement events flow into Server Logs. Whenever one feature unlocks options in another, the UI shows exactly what's available — locked options are marked, never hidden — and every feature still works on its own.

📚 Full documentation Everything is documented in the public wiki — getting started guides, feature deep-dives, command & permission reference, troubleshooting. No account needed, full-text search included. 👉 https://yawbdb.alama.host/wiki

Why YAWBDB? Everything is configurable through a clean web panel — no slash command guessing, no JSON files. Each feature is independent: enable only what you need, configure it visually, and your whole team can manage it without touching the bot.

Bring your own AI keys. No unnecessary permissions.

For the curious YAWBDB is a self-hosted project built around two components: a web panel (PHP backend in laravel, reactive Vue frontend, real-time via WebSockets) and a Discord bot (Node.js, discord.js). They communicate over a private HTTP bridge — the panel evaluates reward graphs and automation logic, the bot executes actions on Discord. The whole thing runs in Docker.

👉 Dashboard: https://yawbdb.alama.host/go/topgg 👉 Docs: https://yawbdb.alama.host/go/topgg-docs