Schedulers & Triggers

Control when and how workflows start execution.

Available Nodes

Node	Type	Purpose
Start	Trigger	Manual workflow start
Cron Scheduler	Trigger (also AI tool)	Run on a recurring schedule
Timer	Action (also AI tool)	Delay execution
Python Executor	Action (also AI tool)	Run Python code

Start

Manually start a workflow with the Run button.

Use Cases

Testing workflows during development
On-demand execution
Entry point for manual processes

Every workflow needs at least one trigger node (Start, Cron Scheduler, a Webhook, or another event trigger) to begin execution.

Cron Scheduler

Run workflows on a recurring schedule. Instead of raw cron expressions, you pick a frequency and then set the matching interval or time-of-day fields. The Cron Scheduler is also a dual-purpose node — it can be connected to an AI Agent’s input-tools handle so agents can schedule delayed or recurring runs programmatically.

Parameters

frequency

select

default:"minutes"

How often the schedule fires: seconds, minutes, hours, days, weeks, months, or once. The other fields below appear based on the frequency you choose.

interval

number

default:"30"

Seconds between fires (5-59). Shown when frequency is seconds.

interval_minutes

number

default:"5"

Minutes between fires (1-59). Shown when frequency is minutes.

interval_hours

number

default:"1"

Hours between fires (1-23). Shown when frequency is hours.

daily_time

select

default:"09:00"

Time of day (HH:MM). Shown when frequency is days.

weekday

select

default:"1"

Day of week for weekly schedules (0=Sunday … 6=Saturday). Shown when frequency is weeks.

weekly_time

select

default:"09:00"

Time of day for weekly schedules. Shown when frequency is weeks.

month_day

select

default:"1"

Day of month (1-28, or L for the last day). Shown when frequency is months.

timezone

select

default:"UTC"

IANA timezone identifier (UTC, America/New_York, America/Los_Angeles, Europe/London, Europe/Berlin, Asia/Tokyo, Asia/Kolkata).

Frequency Examples

Frequency	Fields you set	Result
`minutes`	interval_minutes = 5	Every 5 minutes
`hours`	interval_hours = 1	Every hour
`days`	daily_time = 09:00	Daily at 9am
`weeks`	weekday = 1, weekly_time = 09:00	Every Monday at 9am
`months`	month_day = 1, timezone = UTC	Monthly on the 1st
`once`	—	Fires a single time (no repeat)

Output

{
  "timestamp": "2026-06-15T09:00:00Z",
  "iteration": 1,
  "frequency": "days",
  "timezone": "UTC",
  "schedule": "Daily at 09:00",
  "scheduled_time": "2026-06-15T09:00:00Z",
  "triggered_at": "2026-06-15T09:00:00Z",
  "waited_seconds": 86400,
  "next_run": "Daily at 09:00",
  "message": "Triggered after 1 day, will repeat: Daily at 09:00"
}

Example: Daily Report

[Cron (Daily at 09:00)] --> [HTTP Request] --> [AI Summary] --> [WhatsApp Send]
                            (fetch data)                          (send report)

Timer

Add a delay between nodes. The Timer is a dual-purpose node — it works both as a workflow delay node and as an AI Agent tool.

Parameters

duration

number

default:"1"

Delay duration (1-86400).

unit

select

default:"seconds"

Time unit: seconds, minutes, or hours.

Use Cases

Rate limiting API calls
Waiting for external processes
Scheduling sequential actions
AI Agent tool — agents can set timers programmatically when connected to input-tools

As a Workflow Node

[For Each Item] --> [HTTP Request] --> [Timer (1s)] --> [Next Item]

As an AI Agent Tool

Connect the Timer to an agent’s input-tools handle. The agent can then invoke the timer to introduce delays during reasoning:

[Timer] --(tools handle)--> [AI Agent] --> [Output]

See AI Tools for the full tool schema and LLM interaction details.

Output

{
  "timestamp": "2026-06-15T10:30:05Z",
  "elapsed_ms": 5000,
  "duration": 5,
  "unit": "seconds",
  "message": "Timer completed after 5 seconds"
}

Python Executor

Run custom Python code within workflows.

Parameters

code

required

Python code to execute

Available Variables

Variable	Description
`input_data`	Data from connected input node
`output`	Dictionary to set as node output

Example: Data Processing

# Access input data
message = input_data.get("text", "")

# Process
word_count = len(message.split())
has_question = "?" in message

# Set output
output = {
    "word_count": word_count,
    "has_question": has_question,
    "summary": f"Message has {word_count} words"
}

Example: Conditional Logic

level = input_data.get("battery_level", 100)
is_charging = input_data.get("is_charging", False)

if level < 20 and not is_charging:
    output = {
        "action": "alert",
        "message": f"Low battery: {level}%"
    }
else:
    output = {
        "action": "none"
    }

Example: API Response Formatting

# Parse response (json is available without importing)
data = input_data.get("data", {})
items = data.get("items", [])

# Format for output
formatted = []
for item in items:
    formatted.append({
        "name": item.get("name"),
        "price": f"${item.get('price', 0):.2f}"
    })

output = {
    "count": len(formatted),
    "items": formatted
}

Available Libraries

The Python executor pre-injects a fixed set of names into the namespace:

math - Mathematical functions
json - JSON parsing
datetime, timedelta - Date/time handling
re - Regular expressions
random - Random number generation
Counter, defaultdict - Collections helpers

Reference them directly (for example math.sqrt(4)). import statements are not available in the sandbox; for other modules use the Process Manager node instead.

Code runs in a sandboxed environment. File system and network access are restricted.

Combining Schedulers

Schedule with Delay

[Cron (hourly)] --> [HTTP Request] --> [Timer (5s)] --> [HTTP Request #2]

Multiple Schedules

Deploy workflows with different schedules:

Workflow 1: [Cron (frequency: days)]  --> [Daily Report]
Workflow 2: [Cron (frequency: hours)] --> [Hourly Check]
Workflow 3: [Webhook]                 --> [On-Demand Action]

Workflow States

After deployment, trigger nodes show their state:

State	Color	Description
Idle	Default	Waiting for next trigger
Waiting	Cyan	Actively waiting for event
Running	Purple	Workflow executing

Tips

The Cron Scheduler uses a frequency dropdown — pick minutes/hours/days/weeks/months/once and fill in the matching interval or time-of-day field. No raw cron expression is needed.

Use Python Executor for conditional logic instead of multiple workflow branches.

Add Timer nodes when calling rate-limited APIs repeatedly.

Cron schedules are in UTC by default. Set timezone if needed.

Troubleshooting

Cron not triggering

Verify workflow is deployed (not just saved)
Check that the frequency and its interval/time fields are set
Ensure timezone is correct
Check server logs for errors

Python code fails

Check for syntax errors
Verify input_data has expected fields
Use try/except for error handling
Print debug info with print() (shows in logs)

Timer not working

Ensure the duration is at least 1 (max 86400)
Check the unit (seconds vs minutes vs hours)
Verify node is connected properly

Webhooks

Trigger workflows via HTTP

Message-triggered workflows

​Schedulers & Triggers

​Available Nodes

​Start

​Use Cases

​Cron Scheduler

​Parameters

​Frequency Examples

​Output

​Example: Daily Report

​Timer

​Parameters

​Use Cases

​As a Workflow Node

​As an AI Agent Tool

​Output

​Python Executor

​Parameters

​Available Variables

​Example: Data Processing

​Example: Conditional Logic

​Example: API Response Formatting

​Available Libraries

​Combining Schedulers

​Schedule with Delay

​Multiple Schedules

​Workflow States

​Tips

​Troubleshooting

​Related

Webhooks

WhatsApp

Schedulers & Triggers

Available Nodes

Start

Use Cases

Cron Scheduler

Parameters

Frequency Examples

Output

Example: Daily Report

Timer

Parameters

Use Cases

As a Workflow Node

As an AI Agent Tool

Output

Python Executor

Parameters

Available Variables

Example: Data Processing

Example: Conditional Logic

Example: API Response Formatting

Available Libraries

Combining Schedulers

Schedule with Delay

Multiple Schedules

Workflow States

Tips

Troubleshooting

Related