Airflow Human-in-the-Loop Operators

Implement human approval gates, form inputs, and human-driven branching in Airflow DAGs using the HITL operators. These deferrable operators pause workflow execution until a human responds via the Airflow UI or REST API.

Implementation Checklist

Execute steps in order. Prefer deferrable HITL operators over custom sensors/polling loops.

CRITICAL: Requires Airflow 3.1+. NOT available in Airflow 2.x.

Deferrable: All HITL operators are deferrable—they release their worker slot while waiting for human input.

UI Location: View pending actions at Browse → Required Actions in Airflow UI. Respond via the task instance page's Required Actions tab or the REST API.

Cross-reference: For AI/LLM calls, see the airflow-ai skill.

Step 1: Choose operator

Step 2: Implement operator

ApprovalOperator

from airflow.providers.standard.operators.hitl import ApprovalOperator
from airflow.sdk import dag, task, chain, Param
from pendulum import datetime

@dag(start_date=datetime(2025, 1, 1), schedule="@daily")
def approval_example():
    @task
    def prepare():
        return "Review quarterly report"

    approval = ApprovalOperator(
        task_id="approve_report",
        subject="Report Approval",
        body="{{ ti.xcom_pull(task_ids='prepare') }}",
        defaults="Approve",  # Optional: auto on timeout
        params={"comments": Param("", type="string")},
    )

    @task
    def after_approval(result):
        print(f"Decision: {result['chosen_options']}")

    chain(prepare(), approval)
    after_approval(approval.output)

approval_example()

HITLOperator

Required parameters: subject and options.

from airflow.providers.standard.operators.hitl import HITLOperator
from airflow.sdk import dag, task, chain, Param
from datetime import timedelta
from pendulum import datetime

@dag(start_date=datetime(2025, 1, 1), schedule="@daily")
def hitl_example():
    hitl = HITLOperator(
        task_id="select_option",
        subject="Select Payment Method",
        body="Choose how to process payment",
        options=["ACH", "Wire", "Check"],  # REQUIRED
        defaults=["ACH"],
        multiple=False,
        execution_timeout=timedelta(hours=4),
        params={"amount": Param(1000, type="number")},
    )

    @task
    def process(result):
        print(f"Selected: {result['chosen_options']}")
        print(f"Amount: {result['params_input']['amount']}")

    process(hitl.output)

hitl_example()

HITLBranchOperator

IMPORTANT: Options can either:

1. Directly match downstream task IDs - simpler approach
2. Use options_mapping - for human-friendly labels that map to task IDs

from airflow.providers.standard.operators.hitl import HITLBranchOperator
from airflow.sdk import dag, task, chain
from pendulum import datetime

DEPTS = ["marketing", "engineering", "sales"]

@dag(start_date=datetime(2025, 1, 1), schedule="@daily")
def branch_example():
    branch = HITLBranchOperator(
        task_id="select_dept",
        subject="Select Departments",
        options=[f"Fund {d}" for d in DEPTS],
        options_mapping={f"Fund {d}": d for d in DEPTS},
        multiple=True,
    )

    for dept in DEPTS:
        @task(task_id=dept)
        def handle(dept_name: str = dept):
            # Bind the loop variable at definition time to avoid late-binding bugs
            print(f"Processing {dept_name}")
        chain(branch, handle())

branch_example()

HITLEntryOperator

from airflow.providers.standard.operators.hitl import HITLEntryOperator
from airflow.sdk import dag, task, chain, Param
from pendulum import datetime

@dag(start_date=datetime(2025, 1, 1), schedule="@daily")
def entry_example():
    entry = HITLEntryOperator(
        task_id="get_input",
        subject="Enter Details",
        body="Provide response",
        params={
            "response": Param("", type="string"),
            "priority": Param("p3", type="string"),
        },
    )

    @task
    def process(result):
        print(f"Response: {result['params_input']<