01
Choose a preset or prompt
Repeated work starts with rewriting the same prompt. Create a workspace automation from a preset or your own reusable instructions.
Aiola Automations
Put recurring work
on repeat.
Repeated checks slip when you must remember every prompt and launch. Aiola schedules saved agent work while the app stays open and your desktop stays on. Presets and linked run history make routines reusable and their results reviewable.
How it works
02
Choose execution and schedule
Remembering every launch time interrupts other work. Choose provider, model, Local or Worktree, and schedule; keep Aiola open and the desktop on.
03
Open the resulting thread
An unattended run still needs review. Previous runs links to threads where you can inspect output and errors.
Prompts and execution
Browse reusable presets
Starting routine jobs from scratch takes time. Search presets and saved automations, then filter by source or category to adapt a prompt.
Context mentions
A recurring prompt still needs repository context. Mention files, tasks, agents, logs, feedback, workspaces, issues, and pull requests.
Saved workspace agents
Repeating role instructions makes scheduled prompts harder to maintain. Choose a saved workspace agent to reuse its instructions.
Provider and model
Recurring jobs need access to your preferred agent. Choose Claude Code, Codex, Cursor, Grok CLI, OpenCode, or Antigravity with your own access.
Execution mode and model controls
Recurring work needs the right checkout and model settings. Choose Local or Worktree, plus reasoning, context window, and Fast Mode where supported.
Schedules and run records
Recurring schedule types
Routine checks should fit your working week. Choose daily, weekdays, weekly, or interval schedules and set their timing.
Custom RRULE schedules
Standard schedules may not match your routine. Use a custom RRULE with editable examples for other recurrence patterns.
Manual runs and stopping
You may need a check outside its usual schedule. Run on demand, choose Manual without recurrence, or stop an active run.
Pause and resume
Recurring work sometimes needs a break. Pause an automation and resume its schedule when needed.
App-wide timezone
Inconsistent timezones make recurring work unpredictable. Automations and Calendar use the app-wide timezone configured in Central Settings.
Run history and status
You need to know whether scheduled work actually ran. Inspect next and last run, counts, thread statuses, and completion or failure notifications.
Calendar occurrences
Recurring jobs are hard to compare against task dates. Workspace and Central Calendar show active occurrences that open their automation.
Edit or delete saved routines
Saved routines change as your project evolves. Edit names, icons, prompts, categories, and execution settings, or delete an automation you no longer need.
Context from the previous run
Recurring jobs need a record of what happened before. Aiola includes saved output from the last successful run in the next run’s context.
Schedule a repository review
- Open Automations in the repository workspace and choose a preset.
- Adjust the prompt and add relevant context mentions.
- Choose a provider, model, execution mode, and schedule using the app-wide timezone.
- Keep Aiola open with the desktop on, then inspect the linked run thread.
Problems Aiola solves here
Routine checks depend on remembering to start them. Recurring schedules launch saved prompts while Aiola is open and the desktop is on.
Repeated prompts drift between runs. Presets, saved workspace agents, and context mentions keep instructions available for reuse.
A standard schedule does not fit your routine. Custom RRULE schedules extend daily, weekday, weekly, and interval choices using the app-wide timezone.
An unattended job needs a review trail. Previous runs links to output threads, while run history and notifications expose completion or failure.
A routine needs to stop temporarily. Pause and Resume control recurrence; manual runs and Stop handle individual executions.
Common questions
Build, ship and fix in one app
Errors, user feedback, GitHub and analytics live next to your coding agents in one desktop app. Every issue becomes a task an agent can take.