About Airflow Dev MCP
Airflow Dev MCP is an MCP server in the Automation category: mCP server to facilitate testing Airflow jobs directly from AI coding tools.
Install
git clone https://github.com/BrianLondon/airflow-dev-mcpThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (10)
clear_task_instancesClear tasks so they re-run. Previews as a dry run by default.
Parameters not published.
get_import_errorsShow parse failures with filename and traceback, so you can see why a new DAG isn't showing up.
Parameters not published.
get_run_statusReport a run's state and its per-task states (task, state, try number, operator, timing).
Parameters not published.
get_task_logsReturn the logs for one task attempt, tailed to the last N lines by default.
Parameters not published.
list_connectionsRead Airflow Connections, with passwords omitted.
Parameters not published.
list_dag_runsList recent runs of a DAG, to find a run when you don't have its id.
Parameters not published.
list_dagsList registered DAGs with their paused, active, and import-error flags.
Parameters not published.
list_variablesRead Airflow Variables.
Parameters not published.
set_dag_pausedPause or unpause a DAG. New local DAGs start paused.
Parameters not published.
trigger_dagStart a manual DAG run, optionally with a `conf` payload. Returns the `dag_run_id`.
Parameters not published.
Ask AI
Ask AI about Airflow Dev MCP
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.