mcpField
T
testing-dags

testing-dags

github.com/necatiarslanTesting & quality4446 installsUpdated 21w ago
Repository

About testing-dags

A skill for iteratively testing, debugging, and fixing Airflow DAGs with Airflow VS Code extension tools. It guides users from triggering a run through diagnosis and retesting.

Key features

  1. 01Trigger and monitor DAG runs
  2. 02Analyze failed runs
  3. 03Inspect run details and logs
  4. 04Apply fixes and retest
  5. 05Compare DAG history

Use cases

  1. 01Test a DAG after a code change
  2. 02Diagnose a failed DAG run
  3. 03Review task logs for an error

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/necatiarslan/airflow-vscode-extension --skill testing-dags

After installing, type/testing-dagsin Claude Code to confirm it loaded.

Details

From the project README

DAG Testing Skill

Use the Airflow VS Code extension tools to test, debug, and fix DAGs in iterative cycles.

First Action: Trigger the DAG

When the user asks to test a DAG, your first action should be:

  • trigger_dag_run with the provided DAG ID and config

Do not do pre-flight checks unless the user asks for them.

Testing Workflow Overview

  1. Trigger and monitor
  2. If success, summarize
  3. If failed, debug
  4. Fix and retest

Phase 1: Trigger and Monitor

  1. Trigger: trigger_dag_run
  2. Monitor:
    • Use get_dag_runs to check state
    • If a specific run ID is known, use get_dag_run_detail
Response Interpretation
  • Success: summarize duration and outcome
  • Failed: move to Phase 2
  • Running: ask whether to keep polling

Phase 2: Debug Failures

  1. Use analyse_dag_latest_run for a full analysis of the latest run
  2. If a specific run ID is known, use get_dag_run_detail
  3. If task logs are needed, open go_to_dag_log_view

Phase 3: Fix and Retest

  1. Apply fixes in the DAG or related systems
  2. Re-trigger with trigger_dag_run
  3. Re-check with get_dag_runs or get_dag_run_detail

Notes

  • Prefer analyse_dag_latest_run for fast triage
  • Use get_dag_history when comparing across days

Comments

Markdown supported