# create_cluster

Create a new Databricks cluster with parameters: cluster_name (string, required), spark_version (string, required), node_type_id (string, required), num_workers (integer), autotermination_minutes (integer)

Agent View of the PolicyLayer registry record for `create_cluster`. HTML page: https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/create-cluster

## Facts

- Tool: `create_cluster`
- Server: Databricks MCP Server (`andresgarciasobrado91/databricks-mcp-server`) — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server.md
- Homepage: https://github.com/AndresGarciaSobrado91/databricks-mcp-server
- Risk category: Write (Medium risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "create_cluster",
    "arguments": {}
  }
}
```

## Why create_cluster is rated Medium

This tool creates new computational resources in Databricks, which is a reversible write operation (clusters can be deleted). However, it has high severity because cluster creation incurs infrastructure costs and consumes resources. It does not delete data irreversibly (Destructive), execute arbitrary code (Execute), or move money directly (Financial). It falls squarely into Write: creates infrastructure resources.

From the tool's own definition: "Tool creates a new Databricks cluster with parameters like cluster_name, spark_version, node_type_id, and num_workers. The description explicitly states 'Create a new Databricks cluster', which is a write operation that modifies infrastructure state."

## Use case

AI agents use create_cluster to create or update resources in Databricks MCP Server, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Databricks MCP Server environment.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Databricks MCP Server:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "create_cluster": {
      "limits": [
        {
          "counter": "create_cluster_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Databricks MCP Server (14)

- `terminate_cluster` — Destructive — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/terminate-cluster.md
- `execute_sql` — Execute — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/execute-sql.md
- `repair_run` — Execute — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/repair-run.md
- `run_job` — Execute — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/run-job.md
- `start_cluster` — Execute — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/start-cluster.md
- `export_notebook` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/export-notebook.md
- `get_cluster` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/get-cluster.md
- `get_job` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/get-job.md
- `get_run` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/get-run.md
- `get_run_output` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/get-run-output.md
- `list_clusters` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/list-clusters.md
- `list_files` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/list-files.md
- `list_jobs` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/list-jobs.md
- `list_notebooks` — Read — https://policylayer.com/tools/andresgarciasobrado91-databricks-mcp-server/list-notebooks.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=andresgarciasobrado91-databricks-mcp-server · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/andresgarciasobrado91-databricks-mcp-server