Skill

role-backend:background-jobs

Implements background job processing using Bull/BullMQ (Node.js), Celery (Python), Sidekiq (Ruby), and cron scheduling. Covers job prioritization, retry strategies (exponential backoff), dead letter handling, job monitoring, rate-limited queues, and graceful shutdown. Use when offloading work from request handlers, scheduling tasks, or building job pipelines.

Install

npx claudepluginhub rnavarych/alpha-engineer --plugin role-backend

Tool Access

This skill is limited to using the following tools:

ReadGrepGlobBash

Preview

- Offloading slow or unreliable work from HTTP request handlers

Supporting Assets

references/job-design-retries-dlq.mdreferences/pipelines-shutdown-monitoring.md

SKILL.md

Similar Skills

skill-lookup

Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.

prompts.chat

157.6k

prompt-lookup

Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.

prompts.chat

157.6k

accessibility

Designs, implements, and audits WCAG 2.2 AA accessible UIs for Web (ARIA/HTML5), iOS (SwiftUI traits), and Android (Compose semantics). Audits code for compliance gaps.

everything-claude-code

156.2k

Stats

Parent Repo Stars9

Parent Repo Forks0

Last CommitMar 3, 2026

Actions

View Source View Plugin View on GitHub View README

role-backend:background-jobs

When to use

Offloading slow or unreliable work from HTTP request handlers

Choosing a job queue library for a given language and stack

Designing retry policies for different job types (email, payment, webhook)

Implementing dead letter queues and admin tooling for failed jobs

Building multi-step job pipelines or workflows

Setting up cron/scheduled tasks with distributed lock protection

Configuring graceful worker shutdown for zero-job-loss deploys

Core principles

Idempotency is non-negotiable — every job runs safely multiple times; design for retries from day one

Pass IDs, not data — fetch fresh state inside the job; stale payloads cause subtle bugs

Retries are tiered by criticality — payment jobs retry 5 times over an hour; analytics sync retries 3 times over 15 minutes

DLQ depth is an alert, not a log — non-zero means someone needs to look at it now

Graceful shutdown beats a kill -9 — always drain in-flight jobs before the process exits

Reference Files

references/job-design-retries-dlq.md — technology selection table, idempotency/atomicity/serialization rules, exponential backoff formula, retry config by job type, non-retryable error classification, DLQ setup, job prioritization, and cron scheduling with distributed locks

references/pipelines-shutdown-monitoring.md — BullMQ Flows and Celery chain examples for multi-step workflows, graceful SIGTERM shutdown patterns for Node.js and Python workers, metrics and alerting guidance, and structured log field conventions

Background Jobs

When to use

Offloading slow or unreliable work from HTTP request handlers
Choosing a job queue library for a given language and stack
Designing retry policies for different job types (email, payment, webhook)
Implementing dead letter queues and admin tooling for failed jobs
Building multi-step job pipelines or workflows
Setting up cron/scheduled tasks with distributed lock protection
Configuring graceful worker shutdown for zero-job-loss deploys

Core principles

Idempotency is non-negotiable — every job runs safely multiple times; design for retries from day one
Pass IDs, not data — fetch fresh state inside the job; stale payloads cause subtle bugs
Retries are tiered by criticality — payment jobs retry 5 times over an hour; analytics sync retries 3 times over 15 minutes
DLQ depth is an alert, not a log — non-zero means someone needs to look at it now
Graceful shutdown beats a kill -9 — always drain in-flight jobs before the process exits

Reference Files

references/job-design-retries-dlq.md — technology selection table, idempotency/atomicity/serialization rules, exponential backoff formula, retry config by job type, non-retryable error classification, DLQ setup, job prioritization, and cron scheduling with distributed locks
references/pipelines-shutdown-monitoring.md — BullMQ Flows and Celery chain examples for multi-step workflows, graceful SIGTERM shutdown patterns for Node.js and Python workers, metrics and alerting guidance, and structured log field conventions