Opening your saved workspace. Public pages remain available below.

All playbooks/ Python, Node and cronProject starter kits
FRONTIERS PLAYBOOK / Intermediate

Python, Node and cron

Run your own scripts on a timer, on a server, and know when they break.

YOUR FIRST MOVE

Write a ten-line script that saves one thing to a file, then run it from the command line.

Your learning checklist

0 of 7 steps marked · opening your workspace

STEP 1

Learn cron syntax

Almost every scheduler, from Linux to Vercel to GitHub, uses cron expressions, so learn this once.

docscrontab(5) manual pageDetails +

The reference for time fields, environment variables and shortcuts like @daily.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
toolCrontab.guruDetails +

Type an expression and see it in plain English; Vercel's docs recommend it too.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 2

Write the job in Python or Node.js

Make the job a script you can run by hand first. Then decide who calls it on a schedule.

docsuv documentation (run Python scripts)Details +

Runs a script with its dependencies pinned, so scheduled runs behave like your test runs.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsAPScheduler user guideDetails +

Schedule jobs inside a long-running Python process, with triggers and job stores.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsIntroduction to Node.jsDetails +

Starting point for writing a Node script to schedule.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 3

Schedule it with GitHub Actions (no server)

The cheapest way to run a periodic script: GitHub runs it from your repo and keeps logs. Good for jobs that can run a few minutes late.

docsGitHub Actions: Events that trigger workflows (schedule)Details +

Covers the schedule event, cron syntax and time zone options.

Cost / access
free tier
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsQuickstart for GitHub ActionsDetails +

Create your first workflow and read its run logs.

Cost / access
free tier
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsGitHub Actions billing and usageDetails +

What your plan includes and when charges start; check before scheduling every minute.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 4

Use your host's cron

If your site or service already lives on a platform, its built-in cron is the least maintenance. Both note limits worth knowing.

docsVercel Cron JobsDetails +

Calls a path on your deployment on a schedule; time zone is always UTC and day-of-month with day-of-week cannot both be set.

Cost / access
free tier
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsRailway Cron JobsDetails +

Schedules a service with a crontab expression, with execution rules and timing limits.

Cost / access
unknown
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 5

Run jobs on a VPS with systemd timers and read the logs

On your own server, systemd timers give you logging and retries that plain cron does not. Always know where the logs are.

docssystemd.timer manualDetails +

How .timer files activate a matching .service on a schedule.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsjournalctl manualDetails +

Read and filter logs for any service or timer.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 6

Use a queue when work must not be lost

If a job triggers other jobs or must retry after a failure, a queue is safer than a loop inside one process.

docsBullMQ documentationDetails +

Job and message queue for Node (and other SDKs) with retries and delays, backed by Redis or PostgreSQL.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
docsn8n: Enable queue modeDetails +

A worked example of the same idea inside an automation tool.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
STEP 7

Monitor so you hear when it stops

The dangerous failure is a job that silently stops. Use a 'dead man's switch' that alerts when a ping does not arrive.

docsHealthchecks.io documentationDetails +

Your job pings a URL when it finishes; you get alerted if it is late or missing. Can also be self-hosted.

Cost / access
unknown
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
repoUptime KumaDetails +

A self-hosted monitoring tool for websites and endpoints.

Cost / access
free · Service usage, hosting, or model access may be separate.
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.
repoNetdataDetails +

Server-level monitoring (CPU, memory, disk) for the VPS itself.

Cost / access
unknown
Setup
Follow the linked provider’s current instructions; check account and runtime requirements.

You have made a start when…

  • A script runs on a schedule on a server you control.
  • It logs what it did and alerts you when it fails.
  • You can read its log and find a failure.

Take this into your assistant.

Additional curriculum comes from the supplied AI Frontiers playbooks. Resource descriptions and cost labels are source guidance, not a fresh verification of pricing, compatibility, or results. Open the official instructions before using a tool. Checklist marks are your own progress notes.