### Install Dependencies

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Use this command to install project dependencies with uv.

```bash
uv sync
```

--------------------------------

### Install uv Package Manager

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Instructions for installing the uv package manager. Follow the official documentation for detailed steps.

```bash
# Follow instructions at https://docs.astral.sh/uv/guides/install-python/
```

--------------------------------

### Install Project Dependencies

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Installs project dependencies using the uv package manager after navigating to the project directory.

```bash
cd dremio-mcp-server
uv sync
```

--------------------------------

### File-based Configuration Example

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

An example of a YAML configuration file for Dremio AI settings. It demonstrates how to set Dremio connection details and tool server modes.

```yaml
dremio:
  uri: "https://api.dremio.cloud"
  pat: "@~/tokens/dremio.pat"
  project_id: "project123" # required only for DC
  enable_search: <bool> # Optional: Enable semantic search features
  allow_dml: <bool> # Optional: Allow MCP Server to create views in Dremio
tools:
  server_mode: "FOR_SELF,FOR_DATA_PATTERNS"
```

--------------------------------

### Install Dremio MCP with PAT (Development/Testing)

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Installs the Dremio MCP Helm chart for development or testing using a Personal Access Token (PAT). This is not recommended for production environments.

```bash
helm install dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  --set dremio.pat=your-personal-access-token \
  -n dremio
```

--------------------------------

### Install Dremio MCP with OAuth (Production)

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Installs the Dremio MCP Helm chart in production mode using OAuth authentication. Ensure a namespace named 'dremio' exists.

```bash
kubectl create namespace dremio

helm install dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  -n dremio
```

--------------------------------

### Verify Dremio MCP Server Installation

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Verifies the installation of the Dremio MCP Server by running its help command.

```bash
uv run dremio-mcp-server --help
```

--------------------------------

### Programmatic Configuration with Pydantic

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Demonstrates how to configure Dremio AI settings programmatically using the `dremioai.config.settings` module. Includes loading from a file, getting current settings, and overriding them.

```python
from dremioai.config import settings

# Load from file
settings.configure("path/to/config.yaml")

# Get current settings
current = settings.instance()

# Override settings
current.with_overrides({
    "dremio.uri": "https://new-uri.com",
    "tools.server_mode": "FOR_SELF"
})
```

--------------------------------

### Create a New Dremio Tool

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/tools.md

Example of how to create a new custom tool by inheriting from the `Tools` base class and implementing the `invoke` method.

```python
from typing import ClassVar, Annotated
from dremioai.config.tools import ToolType

class MyNewTool(Tools):
    For: ClassVar[Annotated[ToolType, ToolType.FOR_SELF]]

    async def invoke(self) -> Dict[str, Any]:
        """Tool description here"""
        # Implementation
        pass
```

--------------------------------

### Get Help with Dremio MCP Deployment

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Commands to assist with troubleshooting and understanding the Dremio MCP deployment status.

```bash
kubectl logs -f -l app.kubernetes.io/name=dremio-mcp -n dremio

helm status dremio-mcp -n dremio

kubectl describe all -n dremio
```

--------------------------------

### Define a New Dremio MCP Tool

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Example of how to define a new tool by inheriting from the `Tools` class and implementing the `invoke` method. Ensure to import necessary types.

```python
from dremioai.tools.tools import Tools
from dremioai.config.tools import ToolType

class MyNewTool(Tools):
    For: ClassVar[Annotated[ToolType, ToolType.FOR_SELF]]

    async def invoke(self) -> Dict[str, Any]:
        """Tool description here"""
        # Implementation
        pass

```

--------------------------------

### Dremio MCP Client with SDK (TypeScript)

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/AUTHENTICATION.md

Connects to the MCP server using the official SDK and Dremio token. Requires `@modelcontextprotocol/sdk` to be installed. Use this for client applications to interact with Dremio's Model Context Protocol.

```typescript
// mcp-client.ts - MCP server communication using official SDK
import { Client } from '@modelcontextprotocol/sdk/client/index.js';
import { StreamableHTTPClientTransport } from '@modelcontextprotocol/sdk/client/streamable-http.js';

class DremioMCPClient {
  private mcpServerUrl: string;
  private client: Client | null = null;

  constructor(mcpServerUrl: string) {
    // e.g., "https://dremio-mcp.example.com"
    this.mcpServerUrl = mcpServerUrl;
  }

  /**
   * Connect to MCP server with Dremio token
   * Uses StreamableHTTPClientTransport for compatibility with --enable-streaming-http
   */
  async connect(dremioToken: string): Promise<void> {
    // Create Streamable HTTP transport with authentication
    const transport = new StreamableHTTPClientTransport(
      new URL(this.mcpServerUrl),
      {
        headers: {
          'Authorization': `Bearer ${dremioToken}`
        }
      }
    );

    // Create and connect MCP client
    this.client = new Client({
      name: 'dremio-chat-client',
      version: '1.0.0'
    }, {
      capabilities: {}
    });

    await this.client.connect(transport);
  }

  /**
   * List available tools
   */
  async listTools(): Promise<any> {
    if (!this.client) {
      throw new Error('MCP client not connected');
    }

    return await this.client.listTools();
  }

  /**
   * Close connection
   */
  async close(): Promise<void> {
    if (this.client) {
      await this.client.close();
      this.client = null;
    }
  }
}

```

--------------------------------

### Verify Python Installation for Dremio MCP Server

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Check if the installed Python version meets the requirement of Python 3.11 or later for the Dremio MCP server. This command helps ensure the environment is correctly set up.

```shell
uv python find
```

--------------------------------

### Configure Dremio MCP Server in Claude Desktop

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Add this JSON section to your Claude desktop config file to specify the Dremio MCP server command and arguments. This assumes a local install and picks up the default MCP server config file location.

```json
{
  "globalShortcut": "",
  "mcpServers": {
    "Dremio": {
      "command": "uv",
      "args": [
        "run",
        "--directory", "<toplevel git directory>"
        "dremio-mcp-server",
        "run"
      ]
    }
  }
}
```

--------------------------------

### Get Dremio Application URL

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/templates/NOTES.txt

Use these commands to determine the URL for accessing your Dremio application based on your Kubernetes service configuration.

```bash
{{- if .Values.ingress.enabled }}
{{- range $host := .Values.ingress.hosts }}
  {{- range .paths }}
  http{{ if $.Values.ingress.tls }}s{{ end }}://{{ $host.host }}{{ .path }}
  {{- end }}
{{- end }}
{{- else if contains "NodePort" .Values.service.type }}
  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "dremio-mcp.fullname" . }})
  export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
  echo http://$NODE_IP:$NODE_PORT
{{- else if contains "LoadBalancer" .Values.service.type }}
     NOTE: It may take a few minutes for the LoadBalancer IP to be available.
           You can watch the status of by running 'kubectl get --namespace {{ .Release.Namespace }} svc -w {{ include "dremio-mcp.fullname" . }}'
  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ include "dremio-mcp.fullname" . }} --template "{{"{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}"}})
  echo http://$SERVICE_IP:{{ .Values.service.port }}
{{- else if contains "ClusterIP" .Values.service.type }}
  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/name={{ include "dremio-mcp.name" . }},app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
  export CONTAINER_PORT=$(kubectl get pod --namespace {{ .Release.Namespace }} $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
  echo "Visit http://127.0.0.1:8080 to use your application"
  kubectl --namespace {{ .Release.Namespace }} port-forward $POD_NAME 8080:$CONTAINER_PORT
{{- end }}
```

--------------------------------

### Run MCP Server with Custom Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Launch the MCP server using a specified configuration file. Ensure the config file path is correct.

```bash
uv run dremio-mcp-server run --config-file <path>
```

--------------------------------

### List Available Tools via CLI

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/tools.md

Use this command to list all available tools for a specific mode, such as 'FOR_SELF'.

```bash
uv run dremio-mcp-server tools list -m FOR_SELF
```

--------------------------------

### Create Dremio Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Use this command to create the Dremio configuration file. Provide your Dremio URI and PAT. For Dremio Cloud, include the project ID. The PAT can be provided directly or via a file path prefixed with '@'.

```shell
$ uv run dremio-mcp-server config create dremioai \
    --uri <dremio uri> \
    --pat <dremio pat> \
    # required for cloud: add your project ID if setting up for dremio cloud
    # --project-id <dremio project id>
```

```shell
$ uv run dremio-mcp-server config create dremioai \
    --uri <dremio uri> \
    --pat @/path/to/tokenfile \

```

--------------------------------

### Deploy MCP Server with Helm (Bash)

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/AUTHENTICATION.md

Deploys the MCP server using the provided Helm chart. Ensure Dremio PAT configuration is omitted when using OAuth.

```bash
# For Dremio Software
helm install dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047
```

--------------------------------

### List Default Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

View the default configuration file path and its contents. This command helps in understanding the default settings and their location.

```bash
$ uv run dremio-mcp-server config list
Default config file: /home/user/.config/dremioai/config.yaml (exists = True)
dremio:
  uri: https://api.dremio.cloud
  pat: your-pat-here
  project_id: project123
  experimental_enabled: false
  allow_dml: false
tools:
  server_mode: FOR_SELF
```

--------------------------------

### Show Configuration File Path Only

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Retrieve only the path to the default configuration file without displaying its contents. Useful for scripting or quick checks.

```bash
$ dremio-mcp-server config list --show-filename
Default config file: /home/user/.config/dremioai/config.yaml (exists = True)
```

--------------------------------

### Build and Deploy Dremio MCP Server with Helm

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Build the Docker image and deploy the Dremio MCP server using Helm for production or development/testing. The production deployment uses OAuth, while development uses a Personal Access Token (PAT).

```bash
docker build -t dremio-mcp:0.1.0 .

# Production deployment with OAuth (Recommended)
helm install my-dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047

# Development/Testing with PAT (Not for production)
helm install my-dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  --set dremio.pat=<your-pat>
```

--------------------------------

### List Dremio AI Configurations

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

View all configured Dremio AI connection settings. Useful for verifying or managing multiple configurations.

```bash
uv run dremio-mcp-server config list --type dremioai
```

--------------------------------

### Build Docker Image

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Create a Docker image for the Dremio MCP server. Tag the image with a specific version, e.g., 0.1.0.

```bash
docker build -t dremio-mcp:0.1.0 .
```

--------------------------------

### Create Claude Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

This command creates the configuration file for the Claude LLM desktop app, enabling it to connect with the MCP server.

```shell
$ uv run dremio-mcp-server config create claude
```

--------------------------------

### Run MCP Server

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Execute the MCP server in local stdio mode. For remote deployment, consider streaming HTTP mode.

```bash
uv run dremio-mcp-server run
```

--------------------------------

### Create Dremio AI Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Configure Dremio AI connection details, including URI and Personal Access Token (PAT). This is typically for local or development environments.

```bash
uv run dremio-mcp-server config create dremioai --uri <uri> --pat <pat>
```

--------------------------------

### Dremio MCP Server Project Structure

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Overview of the directory structure for the Dremio MCP Server project, highlighting key directories like src/ and docs/.

```tree
dremio-mcp-server/
├── src/
│   └── dremioai/
│       ├── config/      # Configuration management
│       ├── servers/     # MCP server implementation
│       └── tools/       # Dremio tools implementation
└── docs/                # Documentation

```

--------------------------------

### Environment Variable Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Shows how to configure Dremio AI settings using environment variables. Nested settings are separated by double underscores.

```bash
DREMIOAI_DREMIO__URI="https://api.dremio.cloud"
DREMIOAI_DREMIO__PAT="your-pat-here"
DREMIOAI_TOOLS__SERVER_MODE="FOR_SELF"
```

--------------------------------

### Build Dremio MCP Docker Image

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Builds the Docker image for Dremio MCP. If using a private registry, tag and push the image.

```bash
docker build -t dremio-mcp:0.1.0 .

docker tag dremio-mcp:0.1.0 <your-registry>/dremio-mcp:0.1.0
docker push <your-registry>/dremio-mcp:0.1.0
```

--------------------------------

### Execute a Specific Tool via CLI

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/tools.md

Test a specific tool with arguments using the CLI. Arguments are passed in the format 'arg=value'.

```bash
uv run dremio-mcp-server tools invoke -t RunSqlQuery -c config.yaml \
         args="query=SELECT * FROM sys.nodes"
```

```json
[
  {
    "node_id": "node1",
    "node_type": "COORDINATOR",
    "status": "UP",
    "last_contact": "2024-01-20 10:30:00"
  },
  {
    "node_id": "node2",
    "node_type": "EXECUTOR",
    "status": "UP",
    "last_contact": "2024-01-20 10:29:55"
  }
]
```

--------------------------------

### View Dremio MCP Helm Manifest

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Displays all Kubernetes resources managed by the Dremio MCP Helm release.

```bash
helm get manifest dremio-mcp -n dremio
```

--------------------------------

### View Dremio MCP Helm Values

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Retrieves and displays the current configuration values for the Dremio MCP Helm release.

```bash
helm get values dremio-mcp -n dremio
```

--------------------------------

### Run All Tests

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Execute all project tests using pytest. Strict asyncio mode is enforced.

```bash
uv run pytest tests
```

--------------------------------

### Access Dremio MCP Metrics

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/templates/NOTES.txt

Forward the metrics port to your local machine to access Dremio MCP metrics. This requires metrics to be enabled in your Helm chart values.

```bash
kubectl port-forward --namespace {{ .Release.Namespace }} svc/{{ include "dremio-mcp.fullname" . }} {{ .Values.metrics.port }}:{{ .Values.metrics.port }}
Then visit http://localhost:{{ .Values.metrics.port }}/metrics
```

--------------------------------

### Troubleshoot Dremio MCP Pods

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Commands to help diagnose issues with Dremio MCP pods, including describing pods and checking events.

```bash
kubectl describe pod -l app.kubernetes.io/name=dremio-mcp -n dremio

kubectl get events -n dremio --sort-by='.lastTimestamp'
```

--------------------------------

### List Claude Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Validates the Claude configuration by listing its current settings. Shows the default config file path and its contents.

```shell
$ uv run dremio-mcp-server config list --type claude

Default config file: '/Users/..../Library/Application Support/Claude/claude_desktop_config.json' (exists = True)
{
    'globalShortcut': '',
    'mcpServers': {
        'Dremio': {
            'command': '/opt/homebrew/Cellar/uv/0.6.14/bin/uv',
            'args': [
                'run',
                '--directory',
                '...../dremio-mcp',
                'dremio-mcp-server',
                'run'
            ]
        }
    }
}
```

--------------------------------

### Dremio MCP Server Configuration File Format

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Defines the structure and options for the Dremio MCP server configuration file (config.yaml). Includes settings for Dremio connection, PAT, project ID, search enablement, DML allowance, and server mode.

```yaml
# The dremio section contains 3 main things - the URI to connect, PAT to use
# and optionally the project_id if using with Dremio Cloud
dremio:
    uri: https://.... # the Dremio URI
    pat: "@~/ws/tokens/idl.token" # PAT can be put in a file and used here with @ prefix
    project_id: <string> Project ID required for Dremio Cloud
    enable_search: <bool> # Optional: Enable semantic search
    allow_dml: <bool> # Optional: Allow MCP Server to create views in Dremio
tools:
    server_mode: FOR_DATA_PATTERNS # the serverm

# Optionally the MCP server can also connect and use a prometheus configuration if it
# has been enabled for your Dremio cluster (typically useful for SW installations)
#prometheus:
#uri: ...
#token: ...
```

--------------------------------

### Run Specific Test File

Source: https://github.com/dremio/dremio-mcp/blob/main/CLAUDE.md

Focus testing on a particular file, such as test_chart.py. This is useful for targeted debugging.

```bash
uv run pytest tests/test_chart.py
```

--------------------------------

### Enable Metrics in Dremio MCP

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Upgrades the Dremio MCP deployment to enable metrics collection on a specified port.

```bash
helm upgrade dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  --set metrics.enabled=true \
  --set metrics.port=9091 \
  -n dremio
```

--------------------------------

### Run All Project Tests

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Execute all tests for the project using pytest. This command is typically run from the project's root directory.

```shell
# Run all tests
$ uv run pytest tests
```

--------------------------------

### Check Dremio MCP Logs

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Commands to follow live logs or retrieve logs from a previous, potentially crashed, container for Dremio MCP.

```bash
kubectl logs -f -l app.kubernetes.io/name=dremio-mcp -n dremio

kubectl logs -l app.kubernetes.io/name=dremio-mcp -n dremio --previous
```

--------------------------------

### Prometheus Integration Settings Schema

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Defines the configuration parameters for integrating with a Prometheus server, including its URI and authentication token.

```yaml
prometheus:
  uri: <string|HttpUrl> # Prometheus server URI
  token: <string> # Authentication token
```

--------------------------------

### Dremio Connection Settings Schema

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Defines the structure for Dremio connection settings, including URI, Personal Access Token (PAT), project ID, and options for search and DML.

```yaml
dremio:
  uri: <string|DremioCloudUri> # Dremio instance URI
  pat: <string> # Personal Access Token
  project_id: <string> # Optional: Project ID for Dremio Cloud
  enable_search: <bool> # Optional: Enable semantic search
  allow_dml: <bool> # Optional: Allow MCP Server to create views in Dremio
  wlm:  # Optional: WLM settings, for running MCP server in stdio mode
    engine_name: <string> # Optional: Direct all SQL to this engine if set
```

--------------------------------

### Verify Dremio MCP Deployment

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Commands to check the status of Dremio MCP pods, view logs, and inspect services after deployment.

```bash
kubectl get pods -n dremio

kubectl logs -f -l app.kubernetes.io/name=dremio-mcp -n dremio

kubectl get svc -n dremio
```

--------------------------------

### Enable Autoscaling in Dremio MCP

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Upgrades the Dremio MCP deployment to enable autoscaling with specified replica limits and target CPU utilization.

```bash
helm upgrade dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  --set autoscaling.enabled=true \
  --set autoscaling.minReplicas=2 \
  --set autoscaling.maxReplicas=10 \
  --set autoscaling.targetCPUUtilizationPercentage=80 \
  -n dremio
```

--------------------------------

### Upgrade Dremio MCP Deployment

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Upgrades an existing Dremio MCP deployment using Helm. Ensure the Dremio URI is correctly set.

```bash
helm upgrade dremio-mcp ./helm/dremio-mcp \
  --set dremio.uri=https://dremio.example.com:9047 \
  -n dremio
```

--------------------------------

### Run Dremio MCP Server with Logging Options

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Use these shell commands to run the Dremio MCP server, controlling log file output. You can disable file logging and enable JSON formatted logs, or just enable JSON logging.

```shell
$ uv run dremio-mcp-server run --no-log-to-file --enable-json-logging
```

```shell
# OR 

$ uv run dremio-mcp-server run --enable-json-logging
```

--------------------------------

### List Dremio AI Configuration

Source: https://github.com/dremio/dremio-mcp/blob/main/README.md

Validates the Dremio AI configuration by listing its current settings. Shows the default config file path and its contents, including Dremio URI and PAT.

```shell
$ uv run dremio-mcp-server config list --type dremioai
Default config file: /Users/..../.config/dremioai/config.yaml (exists = True)
dremio:
  enable_search: false
  pat: ....
  uri: ....
tools:
  server_mode: FOR_DATA_PATTERNS
```

--------------------------------

### Tools Settings Schema

Source: https://github.com/dremio/dremio-mcp/blob/main/docs/settings.md

Specifies the configuration for Dremio AI tools, including various server modes like introspection, Prometheus integration, and data pattern analysis.

```yaml
tools:
  server_mode: <string|ToolType|int> # Tool types to enable
```

--------------------------------

### Check Dremio MCP Server Logs

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/templates/NOTES.txt

Stream the logs for your Dremio MCP deployment. This command is useful for debugging and monitoring.

```bash
kubectl logs -f --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/name={{ include "dremio-mcp.name" . }},app.kubernetes.io/instance={{ .Release.Name }}"
```

--------------------------------

### Create a New Feature Branch

Source: https://github.com/dremio/dremio-mcp/blob/main/CONTRIBUTING.md

Creates a new git branch for feature development, branching off the main branch.

```bash
git checkout -b feature/your-feature-name
```

--------------------------------

### Rollback Dremio MCP Deployment

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Rolls back the Dremio MCP deployment to the previous Helm release.

```bash
helm rollback dremio-mcp -n dremio
```

--------------------------------

### Uninstall Dremio MCP Deployment

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/QUICKSTART.md

Uninstalls the Dremio MCP Helm release and associated resources.

```bash
helm uninstall dremio-mcp -n dremio
```

--------------------------------

### Dremio OAuth Token Exchange Service (TypeScript)

Source: https://github.com/dremio/dremio-mcp/blob/main/helm/dremio-mcp/AUTHENTICATION.md

Implements a service to exchange OAuth JWTs for Dremio access tokens and manage token validity. Use this for client-side applications requiring Dremio authentication.

```typescript
// auth.ts - OAuth and token exchange implementation

interface DremioTokenResponse {
  access_token: string;
  token_type: string;
  expires_in: number;
  issued_token_type: string;
  scope: string;
}

class DremioAuthService {
  private dremioBaseUrl: string;
  private dremioToken: string | null = null;
  private tokenExpiry: number = 0;

  constructor(dremioBaseUrl: string) {
    // e.g., "https://dremio.example.com" for Dremio Software
    this.dremioBaseUrl = dremioBaseUrl;
  }

  /**
   * Exchange OAuth JWT for Dremio access token
   * Uses Dremio's OAuth Token Exchange API
   *
   * Reference: https://docs.dremio.com/current/reference/api/oauth-token/#exchanging-an-external-jwt
   */
  async exchangeToken(oauthJwt: string): Promise<DremioTokenResponse> {
    const url = `${this.dremioBaseUrl}/oauth/token`;

    // Prepare form data as per Dremio API specification
    const formData = new URLSearchParams({
      subject_token: oauthJwt,
      subject_token_type: 'urn:ietf:params:oauth:token-type:jwt',
      grant_type: 'urn:ietf:params:oauth:grant-type:token-exchange',
      scope: 'dremio.all'
    });

    const response = await fetch(url, {
      method: 'POST',
      headers: {
        'Content-Type': 'application/x-www-form-urlencoded'
      },
      body: formData.toString()
    });

    if (!response.ok) {
      throw new Error(`Token exchange failed: ${response.status} ${response.statusText}`);
    }

    const tokenData: DremioTokenResponse = await response.json();

    // Store token and expiry
    this.dremioToken = tokenData.access_token;
    this.tokenExpiry = Date.now() + (tokenData.expires_in * 1000);

    return tokenData;
  }

  /**
   * Get valid Dremio token, refreshing if necessary
   */
  async getDremioToken(oauthJwt: string): Promise<string> {
    // Check if token is still valid (with 5 minute buffer)
    if (this.dremioToken && Date.now() < this.tokenExpiry - 300000) {
      return this.dremioToken;
    }

    // Exchange for new token
    const tokenData = await this.exchangeToken(oauthJwt);
    return tokenData.access_token;
  }
}

// Usage in chat application
async function initializeChatSession(oauthJwt: string) {
  const authService = new DremioAuthService(
    'https://dremio.example.com'  // Your Dremio Software instance
  );

  // Exchange OAuth JWT for Dremio token
  const dremioToken = await authService.exchangeToken(oauthJwt);

  // Store token securely (e.g., in session storage with encryption)
  sessionStorage.setItem('dremio_token', dremioToken.access_token);
  sessionStorage.setItem('dremio_token_expiry',
    String(Date.now() + dremioToken.expires_in * 1000));

  return dremioToken.access_token;
}
```

--------------------------------

### Auto-Close Window Script

Source: https://github.com/dremio/dremio-mcp/blob/main/src/dremioai/resources/auth_redirect.html

This JavaScript code automatically closes the browser window after a 3-second delay, providing a seamless user experience after authentication.

```javascript
function closeWindow() {
  window.close();
}

// Try auto-close after 3 seconds
window.onload = function () {
  setTimeout(closeWindow, 3000);
};
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.