Read cron schedules.
Nobody reads cron from memory. You see */15 3 * * 1-5 in a config file, and you either trust the comment above it or paste the whole thing into crontab.guru and hope the site is up. Five stars and some slashes should not require a pilgrimage to a third-party website every time you touch a schedule. This blueprint builds you a decoder that turns any cron expression into a plain sentence and shows exactly when it will fire next, running in your own repository instead of someone else's browser tab.
A cron expression decoder that takes an expression and explains it in plain English, "every 15 minutes past 3am, Monday to Friday," so you can confirm at a glance that it does what the comment claims. Then it predicts the next five run times, so you are not reasoning about schedules in your head. Between the sentence and the concrete timestamps, you catch a wrong field before it ships a job that runs at the wrong hour.
You watch it happen in the task workspace: the agent writes the parsing logic, runs it against sample expressions, and confirms the explanations and predicted times line up. The code lands on a branch in your repository with a pull request ready.
The first version is a base you steer with plain follow-up messages:
Each follow-up wakes the same repository and the agent continues on its own code.
Which cron syntax does it handle? Standard five-field cron out of the box. Ask for six-field or shorthand aliases in a follow-up and the agent extends the parser.
Is it accurate across timezones? It defaults to a clear baseline and can compute next runs in whatever timezone you specify. Just name it in a follow-up.
Where does it run? In your repository, under your account. No pasting internal schedules into a public website to figure out when a job fires.