Standalone Mode

Use standalone mode when you want the fastest setup: one command, one Redis URL, and no app changes.

Run with npx

npx bullstudio -r redis://localhost:6379

Defaults:

Setting	Default
Dashboard URL	`http://localhost:4000`
Redis	`redis://localhost:6379`
Prefix discovery	all prefixes
Dashboard protection	disabled unless a password is set

CLI options

bullstudio \
  --redis redis://localhost:6379 \
  --port 4000 \
  --prefix bull,stage \
  --username operator \
  --password secret \
  --no-open

Option	Short	Description
`--redis <url>`	`-r`	Redis connection URL.
`--port <port>`	`-p`	HTTP port.
`--prefix <prefixes>`		Comma-separated Redis key prefixes. Omit to auto-discover.
`--username <user>`		Dashboard login username when auth is enabled.
`--password <pass>`		Enables dashboard login.
`--no-open`		Do not open the browser automatically.

Environment variables

REDIS_URL=redis://localhost:6379 \
PORT=4000 \
REDIS_PREFIX=bull,stage \
BULLSTUDIO_USERNAME=operator \
BULLSTUDIO_PASSWORD=secret \
bullstudio

Variable	Description
`REDIS_URL`	Redis connection URL.
`PORT`	HTTP port.
`REDIS_PREFIX`	Comma-separated Redis key prefixes.
`BULLSTUDIO_USERNAME`	Login username when `BULLSTUDIO_PASSWORD` is set.
`BULLSTUDIO_PASSWORD`	Enables dashboard login.
`BULLSTUDIO_POLL_ENABLED`	Set to `false` to disable polling entirely.
`BULLSTUDIO_POLL_INTERVAL`	Default poll interval in milliseconds (default `2000`).
`BULLSTUDIO_POLL_MIN_INTERVAL`	Minimum interval in milliseconds operators can select.
`BULLSTUDIO_POLL_ALLOW_OVERRIDE`	Set to `false` to pin the interval and hide the in-dashboard control.

Command-line flags take precedence over environment variables for the same setting.

Standalone mode polls Redis to keep live views current. To ease load on pay-per-command or shared Redis, slow polling down, pin it, or turn it off with the BULLSTUDIO_POLL_* variables above. These mirror the embedded polling config.

# Poll every 10s instead of the 2s default
BULLSTUDIO_POLL_INTERVAL=10000 bullstudio -r redis://localhost:6379

# Disable polling; operators refresh each view manually
BULLSTUDIO_POLL_ENABLED=false bullstudio -r redis://localhost:6379

Operators can still choose their own interval (or Off) from the sidebar, bounded by BULLSTUDIO_POLL_MIN_INTERVAL and BULLSTUDIO_POLL_ALLOW_OVERRIDE.

Docker

Use host.docker.internal when Redis is running on your machine and Bullstudio is running in Docker Desktop:

docker run --rm \
  -p 4000:4000 \
  -e REDIS_URL=redis://host.docker.internal:6379 \
  emirce/bullstudio

Docker Compose:

services:
  bullstudio:
    image: emirce/bullstudio
    ports:
      - "4000:4000"
    environment:
      REDIS_URL: redis://redis:6379
      BULLSTUDIO_USERNAME: operator
      BULLSTUDIO_PASSWORD: secret
    depends_on:
      - redis

  redis:
    image: redis:7-alpine
    ports:
      - "6379:6379"

Redis URLs

# local
bullstudio -r redis://localhost:6379

# password
bullstudio -r redis://:[email protected]:6379

# username and password
bullstudio -r redis://username:[email protected]:6379

# TLS
bullstudio -r rediss://username:[email protected]:6380

Redis Cluster

Cluster mode is detected automatically: point REDIS_URL at any cluster node and Bullstudio connects with a cluster client seeded from that node, discovering the rest of the topology itself. Queue and prefix discovery scans every master node, so hash-tag prefixes (e.g. {orders}) are found wherever their slots live.

# any node of the cluster works as the seed
bullstudio -r redis://cluster-node.example.com:6379

BullMQ requires each queue's keys to share a cluster slot. You can achieve this with either a hash-tagged prefix, such as prefix: "{myprefix}", or a hash-tagged queue name, such as new Queue("{cluster}"). For multiple queues, use distinct hash tags or prefixes so their slots can be distributed across the cluster's masters.

Health check

curl http://localhost:4000/health

{
  "status": "ok",
  "timestamp": "2026-06-01T10:00:00.000Z",
  "redis": "configured"
}

Add a test BullMQ queue

Use this against a local Redis to make the dashboard show data.

import { Queue } from "bullmq";
import IORedis from "ioredis";

const connection = new IORedis("redis://localhost:6379", {
  maxRetriesPerRequest: null,
});

const emailQueue = new Queue("email", { connection });

await emailQueue.add("send-welcome-email", {
  to: "[email protected]",
  template: "welcome",
});

await emailQueue.close();
await connection.quit();

Refresh http://localhost:4000. The queue appears in Overview and Jobs.