Integration · Data orchestration
Apache Airflow + OrchestrAI
Catalog exported 2026-09-02 · Apache Airflow website
Trigger, pause, and re-run Apache Airflow DAGs from chat, with confirmation before each state change.
OrchestrAI exposes 12 Apache Airflow operations: 6 are low-risk (read-only or low-impact), 5 create or modify resources and run only after you confirm the plan, and 1 is destructive and requires a typed risk phrase. 6 of them also carry a step-level approval gate.
What teams use it for
Data teams use OrchestrAI with Airflow to check why a nightly DAG run failed, clear the failed task, and re-trigger it without opening the web UI. Listing DAGs, runs, and task instances is read-only, while triggering, pausing, unpausing, and clearing wait for your confirmation. Deleting a DAG is rated critical and needs a typed risk phrase. There is no operation for editing DAG code, variables, or connections yet, so those changes still happen in your repository or the Airflow console.
Every Apache Airflow operation, with its risk level
| Operation | What it does | Risk | Step-level approval |
|---|---|---|---|
Get Airflow DAG |
Get DAG details | Low risk | No |
Get Airflow DAG Run |
Get DAG run details | Low risk | No |
Get Airflow Task Instance |
Get task instance details | Low risk | No |
List Airflow DAG Runs |
List DAG runs | Low risk | No |
List Airflow DAGs |
List all DAGs in Airflow | Low risk | No |
List Airflow Tasks |
List tasks in a DAG | Low risk | No |
Clear Airflow DAG Run |
Clear a DAG run for re-execution | Creates resources | Yes |
Clear Airflow Task |
Clear a task for re-execution | Creates resources | Yes |
Pause Airflow DAG |
Pause a DAG | Creates resources | Yes |
Trigger Airflow DAG |
Trigger a DAG run | Creates resources | Yes |
Unpause Airflow DAG |
Unpause a DAG | Creates resources | Yes |
Delete Airflow DAG |
Delete a DAG | Destructive | Yes |
Risk tiers come from the catalog: low is read-only or low-impact, medium creates resources and is reversible, high modifies existing resources, destructive may lose data. Every plan that creates or changes resources is shown with its cost estimate and waits for your confirmation. Operations marked with a step-level approval pause again on their own step. Destructive operations require a typed risk phrase.
What you connect
A Apache Airflow credential (stored as airflow).
Connected-service tokens are envelope-encrypted with a per-record key wrapped by a cloud KMS.
Prompts that work
- Show me the last 5 runs of the daily_orders_etl DAG and tell me which tasks failed
- Clear the load_to_warehouse task in run manual__2026-09-01 so it re-executes
- Pause the legacy_sync DAG until I say otherwise
Before anything runs
Every mutation shows its plan, cost estimate, and blast radius, then waits for your confirmation. Destructive operations require a typed risk phrase. Credentials are minted per run through OIDC federation and discarded afterward; nothing you create here is invisible later, because every resource lands in the desired-state ledger where drift is detected and can be converged. Details on the security page.
Frequently asked questions
- Can OrchestrAI trigger an Airflow DAG from chat?
- Yes. Triggering a DAG run is a medium-risk operation, so OrchestrAI shows you the DAG and parameters and waits for your confirmation before calling the Airflow API.
- Will OrchestrAI delete an Airflow DAG by accident?
- No. Deleting a DAG is classified as critical, which means it waits for confirmation and also requires you to type a risk phrase before it runs.
- How does OrchestrAI authenticate to Apache Airflow?
- You add a Apache Airflow credential once in the connections screen. It is envelope-encrypted with a per-record key wrapped by a cloud KMS and is only decrypted inside the run that needs it.
Related integrations
Try it on your own account
Connect your cloud read-only and see your resources, drift, and costs before anything runs. $5 minimum to start. Unused credits refunded in your first 14 days.
Unused credits refunded in your first 14 days.