PgQueuer v1.0 – Python job queue using only PostgreSQ
Ditches Redis entirely by leveraging Postgres LISTEN/NOTIFY for instant wakeups.
A dashboard for ARQ built with FastAPI
TUI and web dashboard for ARQ when the library has no built-in monitoring.
Python developers using ARQ for async job queues
Flower · Celery Flower · RedisInsight
Web dashboard: docker run --rm -p 8000:8000 -e ARQ_DASHBOARD_REDIS_URL=redis://localhost:6379 ghcr.io/srijanpatel/arq-dashboard
TUI: docker run --rm -it -e ARQ_DASHBOARD_REDIS_URL=redis://localhost:6379 ghcr.io/srijanpatel/arq-dashboard tui
Shows per-function runtime percentiles (p50/p95/p99), success rates, throughput (jobs/min), runtime distribution charts, and a visual job timeline from enqueue to completion.
If you're using ARQ and flying blind on what your workers are doing, this might help. Originally forked from ninoseki/arq-dashboard and rewritten with a slightly more modern stack (FastAPI + Vue 3 + Textual) with more stats and a TUI.
Ditches Redis entirely by leveraging Postgres LISTEN/NOTIFY for instant wakeups.
Prometheus alternative in one binary—but early, still rough, needs maturation.
Per-job GPU cost breakdown where cloud bills and nvidia-smi fail to deliver.
Yet another BullMQ dashboard when Arena and Bull Board already exist.
The dashboard exposes cron/interval scheduling, timezone support, retries, execution history, realtime metrics and API-key login, and it runs with a single docker-compose up — exactly the pragmatic feature set you'd want for hosting private webhooks. It isn't reinventing scheduling (Airflow, Rundeck and hosted cron services already exist), but it's a tidy, usable package for teams that want a lightweight, self-hosted alternative with a Python SDK.
Like top for AI agents: tracks token costs across 6 coding agents, 100% offline.