Scheduling & pipelines

Two ways to run workflows without clicking Run each time.

Scheduled runs

Schedules are cron-style, same syntax as Unix cron. Run a workflow daily, weekly, monthly, or any custom cadence.

Creating a schedule

Two entry points:

• Per-agent schedule button (the clock icon on an agent card): schedules the workflow to run with focus on that agent.
• Settings → Automation → + New schedule: schedules the whole workflow.

The Schedule Config popup asks for:

Field	What it does
Name	Display name for the schedule.
Cron expression	Standard cron syntax (e.g. 0 9 * * 1 = every Monday at 9 AM).
Timezone	Defaults to your system timezone; override per schedule if needed.
Workflow file	Path to the .cerevisor to run. Defaults to the currently open workflow.
Run mode	Full workflow, or single-agent focus.
Enabled	Toggle on to activate.

The popup shows:

• Human-readable preview of your cron expression ("Every Monday at 9:00 AM").
• The next 5 fire times so you can sanity-check.

Managing schedules

Settings → Automation lists every schedule. Each row shows:

• Workflow name and schedule name.
• Cron expression and human-readable preview.
• Next fire time.
• Enable/disable toggle.
• Edit and Delete buttons.

What happens when a schedule fires

When the cron expression matches the current time, Cerevisor:

1. Loads the schedule's .cerevisor file fresh from disk.
2. Resolves providers + skills + permissions the same way as a manual run.
3. Runs the workflow.
4. Writes the result to the audit log under a session ID tagged "scheduled".

If Cerevisor isn't running at the scheduled time, the schedule is missed, there's no catch-up run on next launch. (You can change this behavior per schedule if needed.)

License gate

Scheduled runs are a Paid feature. Trial and Free users can create schedules but can't enable them.

Pipelines

A pipeline is an ordered chain of N workflows that run as one unit. Each workflow's output can feed into the next workflow's input.

When to use a pipeline vs. cross-workflow links

• Cross-workflow links (world view): visual, ad-hoc, you see the connections on the canvas.
• Pipelines: structural, ordered, scriptable, easier to schedule.

Use pipelines when the chain is stable and you want to invoke it as one thing. Use cross-workflow links when you're still exploring how the workflows interact.

Creating a pipeline

Title bar → Pipeline icon opens the Pipeline modal.

Configure:

• Steps: ordered list of workflows. Add a step with +; drag rows to reorder.
• Per-step input mapping: for each step (except the first), pick which upstream workflow's output feeds this workflow's input.
• Budget cap: total cost ceiling for the pipeline run. If reached, the run halts mid-pipeline.
• Share skills: toggle: if on, every step inherits the assigned skills from step 1; if off, each step uses its own.
• Share permissions: toggle: if on, every step inherits step 1's WorkflowPermissions.

Click Run pipeline to start.

What happens during a pipeline run

The orchestrator:

1. Loads the first workflow's .cerevisor.
2. Runs it to completion.
3. Extracts its output (the terminal agents' outputs, concatenated).
4. Feeds that output as input to the next workflow's entry agent.
5. Repeats until all steps complete or the budget cap is hit.

The Pipeline run is visible in the Execution Modal with one section per workflow step.

License gate

Pipelines are a Paid feature.

Scheduled pipelines

You can schedule a pipeline. Create the pipeline config, save it as a named pipeline, then create a schedule pointing at the pipeline file (.cerevisor-pipeline).

This is the closest Cerevisor gets to "DAG-of-DAGs" orchestration. It's pretty powerful when you've got it set up, scheduled weekly research → analysis → publish pipelines are the typical case.

Worth knowing

• Schedules and pipelines respect the global WorkflowPermissions settings. If your workflow is set to "ask" for destructive operations, a scheduled run will prompt at run time. For unattended scheduled runs, configure workflow permissions to "allow" the operations the workflow genuinely needs (and audit them via the audit log).
• Cost monitoring matters for unattended runs. Set a per-workflow or per-pipeline budget cap to prevent runaway costs.
• Background-feature providers (Codex CLI, Cursor Agent) work fine for scheduled workflow execution. The scheduler goes through the same dispatch as manual runs.