> ## Documentation Index
> Fetch the complete documentation index at: https://tbd-6fc993ce-hypeship-docker-sandboxes-integration.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Create Kernel App

## `kernel create`

Create a new Kernel application from a template. The CLI provides an interactive prompt to guide you through selecting a language and template, or you can specify all options via flags.

| Flag                          | Description                                                   |
| ----------------------------- | ------------------------------------------------------------- |
| `--name <name>`, `-n`         | Name of the application directory to create.                  |
| `--language <language>`, `-l` | Programming language: `typescript` (`ts`) or `python` (`py`). |
| `--template <template>`, `-t` | Template to scaffold (see available templates below).         |

<Info>If any option is omitted, the CLI will prompt you interactively.</Info>

## Available templates

### TypeScript templates

* **`sample-app`** — Basic template with Playwright integration for page title extraction
* **`captcha-solver`** — Demo of Kernel's auto-CAPTCHA solving capability
* **`anthropic-computer-use`** — Anthropic Computer Use API sample implementation
* **`openai-computer-use`** — OpenAI Computer Using Agent (CUA) sample implementation
* **`gemini-computer-use`** — Google Gemini CUA sample implementation
* **`stagehand`** — Returns the first result of a specified Google search
* **`magnitude`** — [Magnitude framework](https://github.com/magnitude-labs/magnitude) integration

### Python templates

* **`sample-app`** — Basic template with Playwright integration for page title extraction
* **`captcha-solver`** — Demo of Kernel's auto-CAPTCHA solving capability
* **`openagi-computer-use`** — OpenAGI's Lux Computer Use API sample implementation
* **`anthropic-computer-use`** — Anthropic Computer Use API sample implementation
* **`openai-computer-use`** — OpenAI Computer Using Agent (CUA) sample implementation
* **`browser-use`** — Integrates the Browser Use for AI-powered web automation

## Examples

```bash theme={null}
# Interactive mode (prompts for all options)
kernel create

# Create a TypeScript app with the sample template
kernel create --name my-app --language typescript --template sample-app

# Create a Python app with Browser Use
kernel create --name my-scraper --language python --template browser-use

# Create a TypeScript app with Stagehand (shorthand language flag)
kernel create --name my-agent --language ts --template stagehand

# Create a Python Computer Use app
kernel create --name my-cu-app --language py --template anthropic-computer-use
```

## Next steps

After creating your app:

1. **Navigate to the project directory:**
   ```bash theme={null}
   cd my-app
   ```

2. **Install dependencies:**
   * TypeScript: `npm install`
   * Python: `uv venv && source .venv/bin/activate && uv sync`

3. **Authenticate with Kernel:**
   ```bash theme={null}
   kernel login or export KERNEL_API_KEY=<YOUR_API_KEY>
   ```

4. **Deploy your app:**
   ```bash theme={null}
   # TypeScript
   kernel deploy index.ts

   # Python
   kernel deploy main.py
   ```

5. **Invoke your app:**
   ```bash theme={null}
   kernel invoke my-app action-name --payload '{"key": "value"}'
   ```

<Info>Some templates require environment variables (e.g., `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`). Set them using `--env` or `--env-file` flags when deploying. See [`kernel deploy`](/reference/cli/apps#kernel-deploy-file) for details.</Info>