A scheduled chat answers a prompt. A workflow runs your operation.
ChatGPT and Claude can both run a prompt on a timer and send you the result. Cronable is for the work behind the answer: steps that depend on each other, run on a machine you own, and leave a record of every run.
The short answer
A scheduled task in ChatGPT is a saved prompt with a clock. It runs as often as hourly on paid plans, can search the web and use the apps you have connected, and notifies you by push or email. Claude’s Cowork scheduled tasks work much the same way and run in Anthropic’s cloud. ChatGPT also caps how many can be active: 3 on Free, 5 on Plus, 15 on Pro and Enterprise.
Cronable treats the prompt as one step. A workflow fires on a schedule, a webhook, a new file or a new email. Each step runs only if the one before it succeeded, and receives its output; a failure retries, then branches to a handler. Most steps need no model at all — a shell command, an HTTP call, a database query, a message to Slack — and the ones that do run Claude Code, Codex or a model API on your own account.
It runs on your Mac, Windows machine or server, with your files, your network and the command-line tools you are already signed in to. And it is priced by how many jobs you define — 6 on Free, up to 1,000 on Business — never by how many times they run.
Cronable vs ChatGPT scheduled tasks
Where Cronable and ChatGPT scheduled tasks actually differ, capability by capability.
| Capability | Cronable | ChatGPT | Claude Cowork |
|---|---|---|---|
| Where it runs | Your own machine or server, always | OpenAI’s cloud; Work in the desktop app can run tasks on your computer | Anthropic’s cloud, by default |
| Where your workflows and their history live | On your machine — jobs in your own git repo, runs in a local database | In your ChatGPT account | In your Claude account |
| How many you can have | 6 jobs on Free, up to 1,000 on Business; runs never counted | 3 active tasks on Free and Go, 5 on Plus, 10 on Business, 15 on Pro and Enterprise | No cap stated |
| How often | As often as every minute, at exact times, in your timezone | Free: at most daily, in loose time windows. Paid: as often as hourly | Hourly, daily, weekdays or weekly |
| Run a step only if the previous one succeeded | Yes — chain on parent success and pass its output in | Not documented | Not documented |
| Steps that need no model | Shell, SSH, HTTP, SQL, email and 123 integrations | Every task is a model run | Every task is a model run |
| What can start a run | Schedules, parent success, SaaS webhooks, new files, new mail, sheet rows and more | Schedules and monitors; Gmail, Slack and GitHub events with Work, on Plus and up | A schedule, or a manual run |
| Your files, signed-in CLIs and private network | Yes — it runs as you, on your box | Local files and desktop apps you grant to Work in the desktop app | Tasks run remotely; local files and apps only for tasks that run on your computer |
| Run history | Every run, with status, timing, output and live, secret-masked logs | Results on the Scheduled page; each task lives in a chat | Past and upcoming runs on the Scheduled page |
| When a run fails | Retry with backoff, a timeout, a handler job and an alert email | No retry or failure branch documented | No retry or failure branch documented |
| Set it up by describing it | The assistant builds the whole workflow | Ask in chat | Ask in chat |
| Setup effort | A click to install, or one line on a server; a wizard does the rest | None — it is part of ChatGPT | None — it is part of the Claude app |
| Runs while your computer is off | Only if you run it on a server | Yes, for tasks that run in the cloud | Yes, for tasks that run in the cloud |
| What it costs | Free for 6 jobs, then plans sized by jobs; AI usage on your own account | Included with ChatGPT, Free plan included | Included with Pro, Max, Team and Enterprise |
Where it runs
Cronable
Your own machine or server, alwaysChatGPT
OpenAI’s cloud; Work in the desktop app can run tasks on your computerClaude Cowork
Anthropic’s cloud, by defaultWhere your workflows and their history live
Cronable
On your machine — jobs in your own git repo, runs in a local databaseChatGPT
In your ChatGPT accountClaude Cowork
In your Claude accountHow many you can have
Cronable
6 jobs on Free, up to 1,000 on Business; runs never countedChatGPT
3 active tasks on Free and Go, 5 on Plus, 10 on Business, 15 on Pro and EnterpriseClaude Cowork
No cap statedHow often
Cronable
As often as every minute, at exact times, in your timezoneChatGPT
Free: at most daily, in loose time windows. Paid: as often as hourlyClaude Cowork
Hourly, daily, weekdays or weeklyRun a step only if the previous one succeeded
Cronable
Yes — chain on parent success and pass its output inChatGPT
Not documentedClaude Cowork
Not documentedSteps that need no model
Cronable
Shell, SSH, HTTP, SQL, email and 123 integrationsChatGPT
Every task is a model runClaude Cowork
Every task is a model runWhat can start a run
Cronable
Schedules, parent success, SaaS webhooks, new files, new mail, sheet rows and moreChatGPT
Schedules and monitors; Gmail, Slack and GitHub events with Work, on Plus and upClaude Cowork
A schedule, or a manual runYour files, signed-in CLIs and private network
Cronable
Yes — it runs as you, on your boxChatGPT
Local files and desktop apps you grant to Work in the desktop appClaude Cowork
Tasks run remotely; local files and apps only for tasks that run on your computerRun history
Cronable
Every run, with status, timing, output and live, secret-masked logsChatGPT
Results on the Scheduled page; each task lives in a chatClaude Cowork
Past and upcoming runs on the Scheduled pageWhen a run fails
Cronable
Retry with backoff, a timeout, a handler job and an alert emailChatGPT
No retry or failure branch documentedClaude Cowork
No retry or failure branch documentedSet it up by describing it
Cronable
The assistant builds the whole workflowChatGPT
Ask in chatClaude Cowork
Ask in chatSetup effort
Cronable
A click to install, or one line on a server; a wizard does the restChatGPT
None — it is part of ChatGPTClaude Cowork
None — it is part of the Claude appRuns while your computer is off
Cronable
Only if you run it on a serverChatGPT
Yes, for tasks that run in the cloudClaude Cowork
Yes, for tasks that run in the cloudWhat it costs
Cronable
Free for 6 jobs, then plans sized by jobs; AI usage on your own accountChatGPT
Included with ChatGPT, Free plan includedClaude Cowork
Included with Pro, Max, Team and EnterpriseWhat you get by running it on your own machine
Chains that wait for success
A step fires only when its parent succeeded, receives the parent’s output, and a failure branches to a handler job instead of running the rest on broken state. Agent steps sit next to shell, HTTP, database and integration steps in one workflow, on one canvas.
It runs where your work already is
Cronable installs on the machine you actually use — a Mac or Windows app, or a server you own. A job can open your files, use the CLIs you are already signed in to, and reach anything on your network, because it is running as you, on your box.
Your AI agents, on a schedule
Run Claude Code, Codex or Gemini CLI headlessly in a folder on your machine, under the AI account you already use, with a timeout that stops the agent and everything it started. Its final answer becomes the run’s output and flows into the next step.
Every run recorded, with live logs
Each run keeps its status, timing, output and secret-masked logs, streamed live while it happens. Chain a job to fire only when its parent succeeded, retry with backoff, and route a failure to a handler job instead of down the pipe.
Runs are not metered
Plans are sized by how many jobs you define — 6 on Free, up to 1,000 on Business — not by how often they fire. A job that runs every minute costs the same as one that runs monthly. The one thing that is counted per use is the AI assistant, which has a daily message allowance per plan.
Say what you want; it builds the pipeline
Type “every morning, check yesterday’s Stripe refunds and post the total to Slack” and the built-in assistant builds the whole job — the steps, the schedule, the credentials it needs and what happens when one fails — wired and ready to run. You review it on the canvas and change anything. On every plan, including Free.
Cronable and ChatGPT scheduled tasks
Why not just schedule a prompt in ChatGPT?
A scheduled chat answers one prompt on a timer. Most operational work has steps — fetch the numbers, check them, update a record, tell the right person — and each step depends on the last one working. Cronable runs those steps as a chain, waits for each to succeed, hands its output on, and records every run, so you can see exactly where last night’s run stopped.
Can a Cronable workflow still use ChatGPT or Claude?
Yes. A step can call the OpenAI or Anthropic API with your own key, or run Codex or Claude Code headlessly on your machine under the account you already sign in with. The model does the thinking; Cronable decides when it runs, what it gets, and what happens with the answer.
Doesn’t giving it my files make it riskier?
It runs as your user, with your access — the same trust as a script you schedule yourself — so treat it like a shell on the box. The engine listens only on your machine’s loopback address, needs a login for every request, refuses to run as root, and keeps secrets AES-256-GCM encrypted and masked out of every log.
Will results still reach my phone?
Yes. Any step can post to Slack, Teams, Telegram, Discord, SMS or email. From Basic up, remote control lets you check and run jobs from a browser anywhere, through the hosted relay, without opening a port on your network.
Does it run when my laptop is closed?
Only while the machine is awake. A job that fell due during sleep fires once when the machine wakes. For round-the-clock schedules, install Cronable on a server or an always-on machine.
Is there a free plan?
Yes. Free covers 6 jobs, 1 user, 1 machine and 5 AI assistant messages a day, with every built-in job type, git sync and encrypted secrets. Inbound webhooks and remote control start at Basic.
Automate your operations.
Cronable is launching soon. Join the waitlist and we’ll email you the moment self-serve accounts — and the apps for Mac, Windows and Linux — go live.