Astronomer MCP server

What the model can do

Astronomer (Astro) hosted Airflow: workspaces and Deployments with status, Airflow version and executor, code deploys, and on each Airflow 3 Deployment its dags, dag runs, task instances and import errors; when you ask, can trigger a dag run, pause or unpause a dag, re-run (clear) a dag run or chosen tasks, mark a run success or failed, add a note to a run, hibernate or wake a Deployment, and create a workspace.

Reads

• astronomer_list_deploymentsAstro Deployments the token can see: name, workspace, status (HEALTHY, UNHEALTHY, HIBERNATING…), Airflow and Runtime versions, executor…
• astronomer_list_workspacesAstro workspaces: name, description, default cloud and region, and whether CI/CD deploys are enforced by default.
• astronomer_list_deploysCode deploys to one Deployment, newest first as Astro returns them: type (image or dag), status, when, image tag, dag bundle version…
• astronomer_list_dagsDags on one Airflow 3 Deployment: id, display name, paused, schedule, tags, owners, next run, whether it has import errors.
• astronomer_list_dag_runsDag runs on one Airflow 3 Deployment, most recent first: dag, run id, state, type (scheduled, manual…), what triggered it, logical date, start…
• astronomer_list_task_instancesTask instances of one dag run: task, state (success, failed, up_for_retry, upstream_failed…), try number, start, end, duration, operator…
• astronomer_import_errorsDag files on one Airflow 3 Deployment that failed to import, with when and the start of the error, so a dag missing from the list can be explained.

Acts, when you ask

These change something in Astronomer. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• astronomer_trigger_dag_runStarts a new run of a dag on an Astro Deployment now (or for the logical date given)…
• astronomer_set_dag_pausedPauses or unpauses a dag on an Astro Deployment.
• astronomer_clear_dag_runClears a dag run so Airflow runs its tasks again (all tasks, or only the failed ones), re-executing pipeline code.
• astronomer_clear_task_instancesClears the named tasks of one dag run so Airflow runs them again, optionally with their downstream tasks, re-executing pipeline code.
• astronomer_set_dag_run_stateMarks a dag run as success or failed (stopping its running tasks' interest in it), or queued to run again.
• astronomer_set_dag_run_noteSets the note on a dag run, replacing any note it had; the note shows in the Airflow UI to everyone with access.
• astronomer_set_hibernationHibernates an Astro development Deployment (its scheduler and workers stop and no dags run) or wakes it…
• astronomer_end_hibernation_overrideRemoves a Deployment's hibernation override, so it goes back to following its hibernation schedule…
• astronomer_create_workspaceCreates an empty Astro workspace in the organization (no Deployments, no members besides the token's).

Connect it

First connect Astronomer on Dashboard → Connections, then add Memtro to your client once. The Astronomer tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared Astronomer connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Which dag runs failed on the production Deployment last night?
• › Which task failed in this morning's billing_export run?
• › Why is the customer_sync dag missing from the list?
• › What was deployed to the analytics Deployment this week, and from which commit?
• › Re-run the failed tasks of last night's warehouse_load run.

Questions

Is this an official Astronomer MCP server?

No. It is Memtro's MCP server with Memtro's Astronomer connector behind it: the connector reads Astronomer through its API with the credentials you connect on Dashboard → Connections, and Astronomer publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in Astronomer?

Only when you ask, and only what it has been granted: trigger a dag run on a Deployment; pause or unpause a dag; re-run (clear) a dag run; re-run (clear) chosen task instances of a dag run; mark a dag run as success, failed or queued; add or replace the note on a dag run; hibernate or wake a development Deployment; end a Deployment's hibernation override; create a workspace. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server