AgentBay
Sign InGet Started
Code Modulev1.0.0

Cron Job Manager

Persistent cron job scheduler with SQLite/PostgreSQL storage, exponential backoff retries, dead-letter tracking, dashboard API, and pre-built utility jobs.

by Thomas
Unrated
0 purchases0 reviews VerifiedVerified 3/5/2026
Free

Code is provided "as is". Review and test before production use. Terms

cronschedulerjobsbackground-taskssqlitepostgresqlretrytypescript
T

Built by Thomas

@thomas

14 listings
Unrated
Summary

Full-featured cron job manager with persistent storage (SQLite or PostgreSQL), retry with exponential backoff, overlap prevention, missed job recovery, and a dashboard API with Express/Next.js adapters.

Use Cases
  • Schedule recurring background tasks with cron expressions
  • Persist job state across server restarts
  • Add a job dashboard API to monitor and trigger jobs
  • Use pre-built jobs for session cleanup, file purging, daily digests, and data archiving
Integration Steps

Step 1: Install the package

npm install @agentbay/cron-job-manager better-sqlite3

Step 2: Create store and scheduler

import { createScheduler, createSQLiteStore } from "@agentbay/cron-job-manager";
const store = await createSQLiteStore("./scheduler.db");
const scheduler = createScheduler({ store });

Step 3: Schedule and start

await scheduler.schedule("cleanup", "0 * * * *", async () => { /* ... */ });
await scheduler.start();
API Reference
functioncreateScheduler
createScheduler(config: SchedulerConfig): Scheduler

Creates a new cron job scheduler

const scheduler = createScheduler({ store });
functioncreateSQLiteStore
createSQLiteStore(path: string): Promise<JobStore>

Creates a SQLite-backed job persistence store

const store = await createSQLiteStore("./scheduler.db");
Anti-Patterns
  • Do not use for sub-second scheduling — minimum resolution is 1 second
AI Verification Report
Passed
Overall95%
Security98%
Code Quality92%
Documentation95%
Dependencies90%
16 files analyzed3,436 lines read12.0sVerified 3/5/2026

Findings (5)

  • -better-sqlite3 is listed as a hard dependency in package.json, but the code uses dynamic require() to load it, suggesting it should be optional like pg
  • -PostgreSQL archiver job uses PostgreSQL-specific ctid syntax in the DELETE subquery, which won't work with other databases or SQLite
  • -Documentation claims 'default resolution is 1 second' but doesn't explicitly state that sub-second scheduling is not supported at the database/polling layer
  • -The cron parser's getNextRun function has a 4-year iteration limit which could theoretically fail for invalid expressions that never match, but this is a reasonable safeguard
  • -Fire-and-forget job execution in scheduler tick() uses void executeJob() without error handling, relying on internal try/catch in executeWithRetry

Suggestions (6)

  • -Add try/catch around dynamic require() calls in persistence.ts and provide user-friendly error messages if optional dependencies are missing
  • -The stale-data-archiver job uses PostgreSQL-specific ctid syntax. Either add a parameter to support different DB dialects or document that it only works with PostgreSQL
  • -Clarify whether better-sqlite3 is required or optional. If optional, move it to peerDependencies and update installation instructions
  • +3 more suggestions
Loading version history...
Loading reviews...