$ loading_
Inspect Airflow DAGs, runs, logs, and trigger or clear workflows.
This Airflow MCP tool comes from an official registry source and is open source, with no obvious high-risk red flags found. The main considerations are its need for Airflow credentials, its connection to a user-configured Airflow API, and its write-gated administrative actions such as triggering or clearing DAGs.
It requires AIRFLOW_API_URL, AIRFLOW_USERNAME, and AIRFLOW_PASSWORD, which are sensitive credentials for accessing the Airflow control plane. If exposed, they could be used to view DAGs, runs, and logs, and—when write access is enabled—to trigger or clear tasks.
No fixed remote host is listed, but the AIRFLOW_API_URL indicates it will connect to a user-specified Airflow API endpoint. This is routine for its stated function and may transmit management data such as DAGs, run status, task instances, and logs.
The system flags it as executes-code, meaning the MCP at least runs local server-side code. Based on the description, its main capabilities are calling the Airflow API for queries, log access, and—when write-gated is enabled—triggering or clearing runs, with no clear sign of system-level privileges beyond its stated scope.
Its accessible data includes Airflow DAGs, runs, task instances, and logs. If AIRFLOW_ALLOW_WRITE is enabled, it can also modify orchestration state (such as trigger/clear actions). The materials do not indicate a need to read sensitive local files, but access to Airflow data and control-plane functions should be constrained by least privilege.
Positive factors include an official registry source, auditable open-source code, and updates within the past year. However, the lack of a README, no declared license, and very low community adoption (0 stars) reduce external review and maturity signals, so supply-chain posture is best rated as caution rather than high risk.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.us-all/airflow" MCP server from askskill: Run: claude mcp add 'io-github-us-all-airflow' -- npx -y @us-all/airflow-mcp
Connect to Apache Airflow to inspect workflows, trigger DAGs, and monitor health.
Monitor Airflow DAG status, recent runs, history, and performance metrics read-only.
Use a native macOS MCP server to automate system and development tasks securely.
Access Datadog metrics, logs, and alerts for monitoring and incident analysis.
Manage MLflow experiments, runs, model versions, and evaluations in one place.
Query and monitor airspace activity, restrictions, and aviation situational information.