### Deploy and Start App with Databricks Apps CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Sequence of commands to build the wheel, sync the project to a Databricks workspace path, deploy the application by name, and finally start it using the `databricks apps` CLI. Replace placeholders like `<your-profile-name>` and paths.

```bash
uv build --wheel
databricks sync ./build -p <your-profile-name> /Workspace/Users/my-email@org.com/my-app
databricks apps deploy my-app-name -p <your-profile-name> --source-code-path /Workspace/Users/my-email@org.com/my-app
databricks apps start my-app-name -p <your-profile-name>
```

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

### Python Example: Connect and Call Tool

Source: https://github.com/databrickslabs/mcp/blob/master/src/databricks/labs/mcp/static/index.html

An example demonstrating how to connect to a streamable HTTP server using the Databricks SDK and MCP client, initialize a session, and call a tool.

```python
from databricks.sdk import WorkspaceClient
from mcp.client.streamable_http import streamablehttp_client as connect
from mcp import ClientSession

client = WorkspaceClient()

async def main():
    # Connect to a streamable HTTP server

    headers = client.config.authenticate()
    app_url = "https://your.app.url.databricksapps.com/api/mcp/"
    async with connect(app_url, headers=headers) as (
        read_stream,
        write_stream,
        _,
    ):
        # Create a session using the client streams
        async with ClientSession(read_stream, write_stream) as session:
            # Initialize the connection
            await session.initialize()
            # Call a tool
            tool_result = await session.call_tool("echo", {"message": "hello"})

```

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

### Start Local Server with Uvicorn

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Starts the Uvicorn server locally for development, enabling automatic reloads when code changes are detected. This is crucial for rapid iteration during local testing.

```bash
uvicorn custom_server.app:app --reload
```

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

### Configure Databricks Bundle App Command

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Specifies the command to run for the Databricks application within the `app.yaml` configuration file. This tells Databricks how to start the custom server.

```yaml
command: ["uvicorn", "custom_server.app:app"]
```

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

### Run Tests

Source: https://github.com/databrickslabs/mcp/blob/master/CONTRIBUTING.md

Installs requirements using `uv sync` and then runs tests using `uv run pytest tests`.

```bash
uv sync
uv run pytest tests
```

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

### Build, Deploy, and Run with Databricks Bundle

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Builds a wheel package for the application, deploys it to Databricks Apps, and then runs the deployed custom-mcp-server. This sequence is used with the `databricks bundle` CLI.

```bash
uv build --wheel
databricks bundle deploy
databricks bundle run custom-mcp-server
```

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

### Create Databricks App

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Creates a new Databricks App instance named 'mcp-custom-server'. This command provisions the necessary infrastructure on Databricks Apps to host the custom server.

```bash
databricks apps create mcp-custom-server
```

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

### Sync Dependencies with uv

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Synchronizes project dependencies using the `uv` package manager. This is a common step for setting up the local development environment.

```bash
uv sync
```

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

### MCP Server Streamable HTTP URL

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Provides the example URL for connecting to the deployed MCP server using the Streamable HTTP transport. It emphasizes the requirement for the URL to end with `/mcp/`.

```Plaintext
https://your-app-url.usually.ends.with.databricksapps.com/mcp/
```

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

### Configure Databricks Authentication

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Sets the Databricks configuration profile and logs in to authenticate with Databricks. This is a prerequisite for deploying or managing resources via the Databricks CLI.

```bash
export DATABRICKS_CONFIG_PROFILE=<your-profile-name> # e.g. custom-mcp-server
databricks auth login --profile "$DATABRICKS_CONFIG_PROFILE"
```

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

### Configure app.yml for Databricks Apps CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Example `app.yml` configuration file for deploying the UC MCP server using the `databricks apps` CLI. Specifies the Uvicorn command and environment variables.

```yaml
command: ["uvicorn", "databricks.labs.mcp.servers.unity_catalog.app:app"]
env:
  - name: SCHEMA_FULL_NAME
    value: catalog.schema
  - name: GENIE_SPACE_IDS
    value: '["space1","space2"]'
```

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

### Configure app.yml for Databricks Apps Deployment

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Example `app.yml` configuration for deploying an MCP server using `databricks apps` CLI. It specifies the Uvicorn command to run the application and sets required environment variables like `SCHEMA_FULL_NAME` and `GENIE_SPACE_IDS`.

```yaml
command: ["uvicorn", "databricks.labs.mcp.servers.unity_catalog.app:app"]
env:
  - name: SCHEMA_FULL_NAME
    value: catalog.schema
  - name: GENIE_SPACE_IDS
    value: '["space1","space2"]'
```

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

### Sync Source Code and Deploy Databricks App

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Uploads local source code to a specified path in the Databricks workspace and then deploys the Databricks App using the synced code. This command is part of the `databricks apps` CLI deployment workflow.

```bash
DATABRICKS_USERNAME=$(databricks current-user me | jq -r .userName)
databricks sync . "/Users/$DATABRICKS_USERNAME/my-mcp-server"
databricks apps deploy mcp-custom-server --source-code-path "/Workspace/Users/$DATABRICKS_USERNAME/my-mcp-server"
```

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

### Configure MCP Server for Databricks Tools

Source: https://github.com/databrickslabs/mcp/blob/master/src/databricks/labs/mcp/servers/developer_tools/README.md

Example JSON configuration to add the Databricks developer tools MCP server to a client's configuration. This specifies the command and arguments needed to run the server, typically using a package manager like 'uv'.

```json
{
  "mcpServers": {
    "databricks_developer_tools": {
      "command": "/path/to/uv/executable/uv",
      "args": [
        "--directory",
        "/path/to/this/repo/servers/developer_tools",
        "run",
        "databricks-developer-tools"
      ]
    }
  }
}
```

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

### Get Databricks Authentication Token

Source: https://github.com/databrickslabs/mcp/blob/master/src/databricks/labs/mcp/static/index.html

Command to retrieve an authentication token using the Databricks CLI, required for connecting to services.

```bash
databricks auth token -p <your-profile>
```

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

### Get Databricks Auth Token

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Retrieves an authentication token for the Databricks CLI profile, which is required for connecting to deployed Databricks Apps.

```bash
databricks auth token -p your-profile-name
```

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

### Retrieve Databricks Auth Token

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/README.md

Retrieves the authentication token for a specified Databricks profile. This token is used for authenticating API requests, such as connecting to the MCP server.

```bash
databricks auth token -p <name-of-your-profile>
```

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

### Claude Desktop Config for Databricks Unity Catalog MCP Server

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Example JSON configuration for Claude Desktop to connect to a Databricks Unity Catalog MCP server. It specifies the command to run the server and arguments like catalog, schema, and Genie space IDs.

```json
{
  "mcpServers": {
    "databricks_unity_catalog": {
      "command": "/path/to/uv/executable/uv",
      "args": [
        "--directory",
        "/path/to/this/repo",
        "run",
        "unitycatalog-mcp",
        "-s",
        "your_catalog.your_schema",
        "-g",
        "genie_space_id_1,genie_space_id_2"
      ]
    }
  }
}
```

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

### Deploy UC MCP Server using Databricks Apps CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Deploys the UC MCP server on Databricks Apps using the `databricks apps` CLI. This involves building, syncing, deploying, and starting the application.

```bash
uv build --wheel
databricks sync ./build -p <your-profile-name> /Workspace/Users/my-email@org.com/my-app
databricks apps deploy my-app-name -p <your-profile-name> --source-code-path /Workspace/Users/my-email@org.com/my-app
databricks apps start my-app-name -p <your-profile-name>
```

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

### Format and Lint Code

Source: https://github.com/databrickslabs/mcp/blob/master/CONTRIBUTING.md

Formats code using `make fmt` and runs linters using `make lint` from the repository root.

```bash
make fmt
make lint
```

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

### Connect to Unity Catalog MCP Server

Source: https://github.com/databrickslabs/mcp/blob/master/src/databricks/labs/mcp/static/index.html

Details on how to connect to the Unity Catalog MCP server, including transport, URL, and authentication methods.

```APIDOC
Connection Details:
  Transport: Streamable HTTP
  Server URL: https://your.app.url.databricksapps.com/api/mcp/
  Authentication Token:
    Obtain using the Databricks CLI:
    `databricks auth token -p <your-profile>`
  
  Client Connection:
    Use tools like Cloude, MCP Inspector, or other compatible clients.
    Ensure the server URL is set to `https://your.app.url.databricksapps.com/api/mcp/`.
    Use the `Streamable HTTP` Transport with the Authentication header.
```

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

### Build Python Wheel for Databricks Apps Deployment

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Navigates to the project directory and builds the Python wheel package required for deployment with `databricks apps` CLI. This step is crucial before syncing or deploying the application.

```bash
cd /path/to/this/repo
uv build --wheel
```

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

### Build Wheel with uv CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Builds the Python wheel package for the MCP server using the `uv` command-line tool. This is a prerequisite for deploying the server.

```bash
cd /path/to/this/repo
uv build --wheel
```

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

### Connect to MCP Server using Python

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/src/custom_server/static/index.html

Demonstrates how to connect to a streamable HTTP MCP server on Databricks Apps using the Databricks SDK and MCP client libraries. It shows establishing a connection, creating a session, initializing it, and calling a tool.

```python
from databricks.sdk import WorkspaceClient
from databricks_mcp import DatabricksOAuthClientProvider
from mcp.client.streamable_http import streamablehttp_client as connect
from mcp import ClientSession

client = WorkspaceClient()

async def main():
    # Connect to a streamable HTTP server

    app_url = "https://your.app.url.databricksapps.com/mcp/"
    async with connect(app_url, auth=DatabricksOAuthClientProvider(client)) as (
        read_stream,
        write_stream,
        _,
    ):
        # Create a session using the client streams
        async with ClientSession(read_stream, write_stream) as session:
            # Initialize the connection
            await session.initialize()
            # Call a tool
            tool_result = await session.call_tool("echo", {"message": "hello"})

```

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

### Deploy UC MCP Server using Databricks Bundle CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Deploys the UC MCP server on Databricks Apps using the `databricks bundle` CLI. Requires setting environment variables for schema and Genie space IDs.

```bash
BUNDLE_VAR_schema_full_name=catalog.schema BUNDLE_VAR_genie_space_ids=["space1","space2"] \
  databricks bundle deploy -p your-profile-name
```

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

### Format and Lint Python Code

Source: https://github.com/databrickslabs/mcp/blob/master/CONTRIBUTING.md

Runs code formatting and linting tools from the repository root directory to ensure code style consistency and identify potential issues.

```bash
make fmt
make lint
```

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

### Run UC MCP Server App using Databricks Bundle CLI

Source: https://github.com/databrickslabs/mcp/blob/master/README.md

Runs the UC MCP server application on Databricks Apps after deployment using the `databricks bundle` CLI. Ensure environment variables are correctly set.

```bash
BUNDLE_VAR_schema_full_name=catalog.schema BUNDLE_VAR_genie_space_ids=["space1","space2"] \
  databricks bundle run mcp-on-apps -p your-profile-name
```

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

### UV Dependency for Apps CLI Deployment

Source: https://github.com/databrickslabs/mcp/blob/master/examples/custom-server/requirements.txt

This snippet highlights the 'uv' tool as a conditional dependency. It is required specifically when deploying the project using the applications CLI.

```text
uv
```

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