@fluxerjs/core

SDK for building bots on Fluxer.

Install

pnpm add @fluxerjs/core

Usage

Fluxer has no Discord-style gateway intents. Construct a client with no options (or only cache / instance options you need):

import { Client, Events } from '@fluxerjs/core';

const client = new Client();

client.on(Events.Ready, () => console.log('Ready'));
client.on(Events.MessageCreate, async (m) => {
  if (m.content === '!ping') await m.reply('Pong');
});

await client.login(process.env.FLUXER_BOT_TOKEN);

Documentation

Full guides, examples, REST reference, and SDK API docs: https://fluxer.js.org

Start with Installation → Basic bot → Prefix commands → Errors → Caching.

Coming from discord.js? Read From discord.js. Upgrading from 2.x? See Upgrading to 3.0. Coming from 1.x? See Migrating to 2.0 first.

From this repo:

# Dev server: http://localhost:3333
pnpm run docs:dev

# Build packages + generate docs + Next build
pnpm run docs:build

# Preview the production build
pnpm run docs:preview

• docs:dev: Next.js on port 3333
• docs:build: Generates API docs JSON, then builds the Next.js docs site (via turbo)
• docs:preview: next start on port 3333

Next steps

• Runnable bots: examples/ (copy examples/.env.example)
• Caching when you care about memory and identity
• Collectors, permissions, embeds

Advanced: cluster and sharding

One Client = one Fluxer instance (token, REST, gateway, caches, CDN/invite URLs). Each instance needs its own bot token.

Beta: ClientCluster can add/remove/restart runtimes without restarting the process. The API may change; constructing a cluster emits a FluxerClientClusterBeta warning. restart(id, { token }) requires re-supplying the instance token (never stored).

import { ClientCluster, ClientClusterEvents, Events } from '@fluxerjs/core';
// or: import { ClientCluster } from '@fluxerjs/core/cluster';

const cluster = new ClientCluster({
  configure(runtime) {
    runtime.client.on(Events.MessageCreate, async (m) => {
      if (m.content === '!ping') await m.reply(`Pong from ${runtime.id}`);
    });
  },
});

await cluster.add({ id: 'hosted', token: process.env.FLUXER_BOT_TOKEN });
await cluster.add({
  id: 'self',
  token: process.env.SELFHOST_BOT_TOKEN, // must be issued by the self-hosted instance
  discovery: 'https://api.my.instance',
});

// Later: no process restart
await cluster.restart('self', { token: process.env.SELFHOST_BOT_TOKEN });
await cluster.remove('self');

You can still manage raw Client instances yourself. See examples/multi-instance-bot.js and the multi-instance guide.

For process-per-shard scaling, see @fluxerjs/sharding and examples/sharded-bot.js.

License

Apache-2.0