# Configure apps
> This bundle contains all pages in the Configure apps section.
> Source: https://www.union.ai/docs/v2/union/user-guide/configure-apps/

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps ===

# Configure apps

> **📝 Note**
>
> An LLM-optimized bundle of this entire section is available at [`section.md`](section.md).
> This single file contains all pages in this section, optimized for AI coding agent context.

`[[AppEnvironment]]`s allows you to configure the environment in which your app runs, including the container image, compute resources, secrets, domains, scaling behavior, and more.

Similar to `[[TaskEnvironment]]`, configuration can be set when creating the `[[AppEnvironment]]` object. Unlike tasks, apps are long-running services, so they have additional configuration options specific to web services:

- `port`: What port the app listens on
- `command` and `args`: How to start the app
- `scaling`: Autoscaling configuration for handling variable load
- `domain`: Custom domains and subdomains for your app
- `requires_auth`: Whether the app requires authentication to access
- `depends_on`: Other app or task environments that the app depends on

## Hello World example

Here's a complete example of deploying a simple Streamlit "hello world" app with a custom subdomain.

There are two ways to build apps in Flyte:
1. Defining `AppEnvironment(.., args=[...])` to run the app with the underlying `fserve` command.
2. Defining `@app_env.server` to run the app with a custom server function.

### Using fserve args

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment image}}
image = flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages("streamlit==1.41.1")
# {{/docs-fragment image}}

# {{docs-fragment app-env}}
app_env = flyte.app.AppEnvironment(
    name="hello-world-app",
    image=image,
    args=["streamlit", "hello", "--server.port", "8080"],
    port=8080,
    resources=flyte.Resources(cpu="1", memory="1Gi"),
    requires_auth=False,
    domain=flyte.app.Domain(subdomain="hello"),
)
# {{/docs-fragment app-env}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    flyte.init_from_config()

    # Deploy the app
    app = flyte.serve(app_env)
    print(f"App served at: {app.url}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/hello-world-app.py*

This example demonstrates:

- Creating a custom Docker image with Streamlit
- Setting the `args` to run the Streamlit hello app, which uses the underlying `fserve` command to run the app.
- Configuring the port
- Setting resource limits
- Disabling authentication (for public access)
- Using a custom subdomain

### Using @app_env.server

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment image}}
image = flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages("streamlit==1.41.1")
# {{/docs-fragment image}}

# {{docs-fragment app-env}}
app_env = flyte.app.AppEnvironment(
    name="hello-world-app-server",
    image=image,
    port=8080,
    resources=flyte.Resources(cpu="1", memory="1Gi"),
    requires_auth=False,
    domain=flyte.app.Domain(subdomain="hello-server"),
)

@app_env.server
def server():
    import subprocess
    subprocess.run(["streamlit", "hello", "--server.port", "8080"], check=False)
# {{/docs-fragment app-env}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    flyte.init_from_config()

    # Deploy the app
    app = flyte.serve(app_env)
    print(f"App served at: {app.url}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/hello-world-app-server.py*

This example demonstrates:

- Creating a custom Docker image with Streamlit
- Using the `@app_env.server` decorator to define a server function that runs the Streamlit hello app.
- Configuring the port
- Setting resource limits
- Disabling authentication (for public access)
- Using a custom subdomain

Once deployed, your app will be accessible at the generated URL or your custom subdomain.

## Differences from TaskEnvironment

While `AppEnvironment` inherits from `Environment` (the same base class as `TaskEnvironment`), it has several app-specific parameters:

| Parameter | AppEnvironment | TaskEnvironment | Description |
|-----------|----------------|-----------------|-------------|
| `type` | ✅ | ❌ | Type of app (e.g., "FastAPI", "Streamlit") |
| `port` | ✅ | ❌ | Port the app listens on |
| `args` | ✅ | ❌ | Arguments to pass to the app |
| `command` | ✅ | ❌ | Command to run the app |
| `requires_auth` | ✅ | ❌ | Whether app requires authentication |
| `scaling` | ✅ | ❌ | Autoscaling configuration |
| `domain` | ✅ | ❌ | Custom domain/subdomain |
| `links` | ✅ | ❌ | Links to include in the App UI page |
| `include` | ✅ | ❌ | Files to include in app |
| `parameters` | ✅ | ❌ | Parameters to pass to app |
| `cluster_pool` | ✅ | ❌ | Cluster pool for deployment |

Parameters like `image`, `resources`, `secrets`, `env_vars`, and `depends_on` are shared between both environment types. See the [task configuration](https://www.union.ai/docs/v2/union/user-guide/task-configuration/_index) docs for details on these shared parameters.

## Configuration topics

Learn more about configuring apps:

- **Configure apps > App environment settings**: Images, resources, secrets, and app-specific settings like `type`, `port`, `args`, `requires_auth`
- **Configure apps > App environment settings**: Understanding the difference between `args` and `command`
- **Configure apps > Including additional files**: How to include additional files needed by your app
- **Configure apps > Passing parameters into app environments**: Pass parameters to your app at deployment time
- **Configure apps > App environment settings**: Configure scaling up and down based on traffic with idle TTL
- **Configure apps > Apps depending on other environments**: Use `depends_on` to deploy dependent apps together

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps/app-environment-settings ===

# App environment settings

`[[AppEnvironment]]`s control how your apps run in Flyte, including images, resources, secrets, startup behavior, and autoscaling.

## Shared environment settings

`[[AppEnvironment]]`s share many configuration options with `[[TaskEnvironment]]`s:

- **Images**: See [Container images](https://www.union.ai/docs/v2/union/user-guide/task-configuration/container-images) for details on creating and using container images
- **Resources**: See [Resources](https://www.union.ai/docs/v2/union/user-guide/task-configuration/resources) for CPU, memory, GPU, and storage configuration
- **Secrets**: See [Secrets](https://www.union.ai/docs/v2/union/user-guide/task-configuration/secrets) for injecting secrets into your app
- **Environment variables**: Set via the `env_vars` parameter (same as tasks)
- **Cluster pools**: Specify via the `cluster_pool` parameter

## App-specific environment settings

For complete parameter documentation and type signatures, see the [`AppEnvironment` API reference](https://www.union.ai/docs/v2/union/api-reference/flyte-sdk/packages/flyte.app/appenvironment).

### `type`

The `type` parameter is an optional string that identifies what kind of app this is. It's used for organizational purposes and may be used by the UI or tooling to display or filter apps.

```python
app_env = flyte.app.AppEnvironment(
    name="my-fastapi-app",
    type="FastAPI",
    # ...
)
```

When using specialized app environments like `FastAPIAppEnvironment`, the type is automatically set. For custom apps, you can set it to any string value.

### `port`

The `port` parameter specifies which port your app listens on. It can be an integer or a `Port` object.

```python
# Using an integer (simple case)
app_env = flyte.app.AppEnvironment(name="my-app", port=8080, ...)

# Using a Port object (more control)
app_env = flyte.app.AppEnvironment(
    name="my-app",
    port=flyte.app.Port(port=8080),
    # ...
)
```

The default port is `8080`. Your app should listen on this port (or the port you specify).

> [!NOTE]
> Ports 8012, 8022, 8112, 9090, and 9091 are reserved and cannot be used for apps.

### `args`

The `args` parameter specifies arguments to pass to your app's command. This is typically used when you need to pass additional arguments to the command specified in `command`, or when using the default command behavior.

```python
app_env = flyte.app.AppEnvironment(
    name="streamlit-app",
    args="streamlit run main.py --server.port 8080",
    port=8080,
    # ...
)
```

`args` can be either a string (which will be shell-split) or a list of strings:

```python
# String form (will be shell-split)
args="--option1 value1 --option2 value2"

# List form (more explicit)
args=["--option1", "value1", "--option2", "value2"]
```

#### Environment variable substitution

Environment variables are automatically substituted in `args` strings when they start with the `$` character. This works for both:

- Values from `env_vars`
- Secrets that are specified as environment variables (via `as_env_var` in `flyte.Secret`)

The `$VARIABLE_NAME` syntax will be replaced with the actual environment variable value at runtime:

```python
# Using env_vars
app_env = flyte.app.AppEnvironment(
    name="my-app",
    env_vars={"API_KEY": "secret-key-123"},
    args="--api-key $API_KEY",  # $API_KEY will be replaced with "secret-key-123"
    # ...
)

# Using secrets
app_env = flyte.app.AppEnvironment(
    name="my-app",
    secrets=flyte.Secret(key="AUTH_SECRET", as_env_var="AUTH_SECRET"),
    args=["--api-key", "$AUTH_SECRET"],  # $AUTH_SECRET will be replaced with the secret value
    # ...
)
```

This is particularly useful for passing API keys or other sensitive values to command-line arguments without hardcoding them in your code. The substitution happens at runtime, ensuring secrets are never exposed in your code or configuration files.

> [!TIP]
> For most `AppEnvironment`s, use `args` instead of `command` to specify the app startup command
> in the container. This is because `args` will use the `fserve` command to run the app, which
> unlocks features like local code bundling and file/directory mounting via parameter injection.

### `command`

The `command` parameter specifies the full command to run your app. If not specified, Flyte will use a default command that runs your app via `fserve`, which is the Python executable provided
by `flyte` to run apps.

```python
# Explicit command
app_env = flyte.app.AppEnvironment(
    name="streamlit-hello",
    command="streamlit hello --server.port 8080",
    port=8080,
    # ...
)

# Using default command (recommended for most cases)
# When command is None, Flyte generates a command based on your app configuration
app_env = flyte.app.AppEnvironment(name="my-app", ...)  # command=None by default
```

> [!TIP]
> For most apps, especially when using specialized app environments like `FastAPIAppEnvironment`, you don't need to specify `command` as it's automatically configured. Use `command` when you need
> to specify the raw container command, e.g. when running a non-Python app or when you have all
> of the dependencies and data used by the app available in the container.

### `requires_auth`

The `requires_auth` parameter controls whether the app requires authentication to access. By default, apps require authentication (`requires_auth=True`).

```python
# Public app (no authentication required)
app_env = flyte.app.AppEnvironment(
    name="public-dashboard",
    requires_auth=False,
    # ...
)

# Private app (authentication required - default)
app_env = flyte.app.AppEnvironment(
    name="internal-api",
    requires_auth=True,
    # ...
)  # Default
```

When `requires_auth=True`, users must authenticate with Flyte to access the app. When `requires_auth=False`, the app is publicly accessible (though it may still require API keys or other app-level authentication).

### `domain`

The `domain` parameter specifies a custom domain or subdomain for your app. Use `flyte.app.Domain` to configure a subdomain or custom domain.

```python
app_env = flyte.app.AppEnvironment(
    name="my-app",
    domain=flyte.app.Domain(subdomain="myapp"),
    # ...
)
```

### `links`

The `links` parameter adds links to the App UI page. Use `flyte.app.Link` objects to specify relative or absolute links with titles.

```python
app_env = flyte.app.AppEnvironment(
    name="my-app",
    links=[
        flyte.app.Link(path="/docs", title="API Documentation", is_relative=True),
        flyte.app.Link(path="/health", title="Health Check", is_relative=True),
        flyte.app.Link(path="https://www.example.com", title="External link", is_relative=False),
    ],
    # ...
)
```

### `include`

The `include` parameter specifies files and directories to include in the app bundle. Use glob patterns or explicit paths to include code files needed by your app.

```python
app_env = flyte.app.AppEnvironment(
    name="my-app",
    include=["*.py", "models/", "utils/", "requirements.txt"],
    # ...
)
```

> [!NOTE]
> Learn more about including additional files in your app deployment [here](./including-additional-files).

### `parameters`

The `parameters` parameter passes parameters to your app at deployment time. Parameters can be primitive values, files, directories, or delayed values like `RunOutput` or `AppEndpoint`.

```python
app_env = flyte.app.AppEnvironment(
    name="my-app",
    parameters=[
        flyte.app.Parameter(name="config", value="foo", env_var="BAR"),
        flyte.app.Parameter(name="model", value=flyte.io.File(path="s3://bucket/model.pkl"), mount="/mnt/model"),
        flyte.app.Parameter(name="data", value=flyte.io.File(path="s3://bucket/data.pkl"), mount="/mnt/data"),
    ],
    # ...
)
```

> [!NOTE]
> Learn more about passing parameters to your app at deployment time [here](./passing-parameters).

### `scaling`

The `scaling` parameter configures autoscaling behavior for your app. Use `flyte.app.Scaling` to set replica ranges and idle TTL.

```python
app_env = flyte.app.AppEnvironment(
    name="my-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 5),
        scaledown_after=300,  # Scale down after 5 minutes of idle time
    ),
    # ...
)
```

> [!NOTE]
> Learn more about autoscaling apps [here](./auto-scaling-apps).

### `depends_on`

The `depends_on` parameter specifies environment dependencies. When you deploy an app, all dependencies are deployed first.

```python
backend_env = flyte.app.AppEnvironment(name="backend-api", ...)

frontend_env = flyte.app.AppEnvironment(
    name="frontend-app",
    depends_on=[backend_env],  # backend-api will be deployed first
    # ...
)
```

> [!NOTE]
> Learn more about app environment dependencies [her e](./apps-depending-on-environments).

## App startup

There are two ways to start up an app in Flyte:
1. With a server function using `@app_env.server`
2. As a container command using `command` or `args`

### Server decorator via `@app_env.server`

The server function is a Python function that runs the app. It is defined using the `@app_env.server` decorator.

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "uvicorn",
#    "flyte>=2.0.0b52",
# ]
# ///

import fastapi
import uvicorn

import flyte
from flyte.app.extras import FastAPIAppEnvironment

# {{docs-fragment fastapi-app}}
app = fastapi.FastAPI()

env = FastAPIAppEnvironment(
    name="configure-fastapi-example",
    app=app,
    image=flyte.Image.from_uv_script(__file__, name="configure-fastapi-example"),
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
    port=8080,
)

@env.server
def server():
    print("Starting server...")
    uvicorn.run(app, port=8080)

@app.get("/")
async def root() -> dict:
    return {"message": "Hello from FastAPI!"}
# {{/docs-fragment fastapi-app}}

# {{docs-fragment on-startup-decorator}}
state = {}

@env.on_startup
async def app_startup():
    print("App started up")
    state["data"] = ["Here's", "some", "data"]
# {{/docs-fragment on-startup-decorator}}

# {{docs-fragment on-shutdown-decorator}}
@env.on_shutdown
async def app_shutdown():
    print("App shut down")
    state.clear()  # clears the data
# {{/docs-fragment on-shutdown-decorator}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    import logging

    flyte.init_from_config(log_level=logging.DEBUG)
    deployed_app = flyte.serve(env)
    print(f"App served at: {deployed_app.url}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/fastapi-server-example.py*

The `@app_env.server` decorator allows you to define a synchronous or asynchronous function that runs the app, either
with a server start command like `uvicorn.run`, [`HTTPServer.serve_forever`](https://docs.python.org/3/library/http.server.html), etc.

> [!NOTE]
> Generally the `[[FastAPIAppEnvironment]]` handles serving automatically under the hood,
> the example above just shows how the `@app_env.server` decorator can be used to define a server function
> that runs the app.

#### Startup hook

The server function is called after the app is started up, and before the app is shut down. It is defined using the `@app_env.on_startup` decorator. This is useful if you need to load any state or external connections needed to run the
app before it starts.

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "uvicorn",
#    "flyte>=2.0.0b52",
# ]
# ///

import fastapi
import uvicorn

import flyte
from flyte.app.extras import FastAPIAppEnvironment

# {{docs-fragment fastapi-app}}
app = fastapi.FastAPI()

env = FastAPIAppEnvironment(
    name="configure-fastapi-example",
    app=app,
    image=flyte.Image.from_uv_script(__file__, name="configure-fastapi-example"),
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
    port=8080,
)

@env.server
def server():
    print("Starting server...")
    uvicorn.run(app, port=8080)

@app.get("/")
async def root() -> dict:
    return {"message": "Hello from FastAPI!"}
# {{/docs-fragment fastapi-app}}

# {{docs-fragment on-startup-decorator}}
state = {}

@env.on_startup
async def app_startup():
    print("App started up")
    state["data"] = ["Here's", "some", "data"]
# {{/docs-fragment on-startup-decorator}}

# {{docs-fragment on-shutdown-decorator}}
@env.on_shutdown
async def app_shutdown():
    print("App shut down")
    state.clear()  # clears the data
# {{/docs-fragment on-shutdown-decorator}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    import logging

    flyte.init_from_config(log_level=logging.DEBUG)
    deployed_app = flyte.serve(env)
    print(f"App served at: {deployed_app.url}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/fastapi-server-example.py*

#### Shutdown hook

The server function is called before the app instance shuts down during scale down. It is defined using the
`@app_env.on_shutdown` decorator. This is useful if you need to clean up any state or external connections in the
container running the app.

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "uvicorn",
#    "flyte>=2.0.0b52",
# ]
# ///

import fastapi
import uvicorn

import flyte
from flyte.app.extras import FastAPIAppEnvironment

# {{docs-fragment fastapi-app}}
app = fastapi.FastAPI()

env = FastAPIAppEnvironment(
    name="configure-fastapi-example",
    app=app,
    image=flyte.Image.from_uv_script(__file__, name="configure-fastapi-example"),
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
    port=8080,
)

@env.server
def server():
    print("Starting server...")
    uvicorn.run(app, port=8080)

@app.get("/")
async def root() -> dict:
    return {"message": "Hello from FastAPI!"}
# {{/docs-fragment fastapi-app}}

# {{docs-fragment on-startup-decorator}}
state = {}

@env.on_startup
async def app_startup():
    print("App started up")
    state["data"] = ["Here's", "some", "data"]
# {{/docs-fragment on-startup-decorator}}

# {{docs-fragment on-shutdown-decorator}}
@env.on_shutdown
async def app_shutdown():
    print("App shut down")
    state.clear()  # clears the data
# {{/docs-fragment on-shutdown-decorator}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    import logging

    flyte.init_from_config(log_level=logging.DEBUG)
    deployed_app = flyte.serve(env)
    print(f"App served at: {deployed_app.url}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/fastapi-server-example.py*

### Container command via `command` vs `args`

The difference between `args` and `command` is crucial for properly configuring how your app starts.

- **`command`**: The full command to run your app, for example, `"streamlit hello --server.port 8080"`. For most use
  cases, you don't need to specify `command` as it's automatically configured, and uses the `fserve` executable to
  run the app. `fserve` does additional setup for you, like setting up the code bundle and loading [parameters](./passing-parameters) if provided, so it's highly recommended to use the default command.
- **`args`**: Arguments to pass to your app's command (used with the default Flyte command or your custom command). The
  `fserve` executable takes in additional arguments, which you can specify as the arguments needed to run your app, e.g.
  `uvicorn run main.py --server.port 8080`.

#### Default startup behavior

When you don't specify a `command`, Flyte generates a default command that uses `fserve` to run your app. This default command handles:

- Setting up the code bundle
- Configuring the version
- Setting up project/domain context
- Injecting parameters if provided

The default command looks like:

```bash
fserve --version <version> --project <project> --domain <domain> -- <args>
```

So if you specify `args`, they'll be appended after the `--` separator.

#### Using args with the default command

When you use `args` without specifying `command`, the args are passed to the default Flyte command:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment args-with-default-command}}
# Using args with default command
app_env = flyte.app.AppEnvironment(
    name="streamlit-app",
    args="streamlit run main.py --server.port 8080",
    port=8080,
    include=["main.py"],
    # command is None, so default Flyte command is used
)
# {{/docs-fragment args-with-default-command}}

# {{docs-fragment explicit-command}}
# Using explicit command
app_env2 = flyte.app.AppEnvironment(
    name="streamlit-hello",
    command="streamlit hello --server.port 8080",
    port=8080,
    # No args needed since command includes everything
)
# {{/docs-fragment explicit-command}}

# {{docs-fragment command-with-args}}
# Using command with args
app_env3 = flyte.app.AppEnvironment(
    name="custom-app",
    command="python -m myapp",
    args="--option1 value1 --option2 value2",
    # This runs: python -m myapp --option1 value1 --option2 value2
)
# {{/docs-fragment command-with-args}}

# {{docs-fragment fastapi-auto-command}}
# FastAPIAppEnvironment automatically sets command
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

app = FastAPI()

env = FastAPIAppEnvironment(
    name="my-api",
    app=app,
    # You typically don't need to specify command or args, since the
    # FastAPIAppEnvironment automatically uses the bundled code to serve the
    # app via uvicorn.
)
# {{/docs-fragment fastapi-auto-command}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/app-startup-examples.py*

This effectively runs:

```bash
fserve --version ... --project ... --domain ... -- streamlit run main.py --server.port 8080
```

#### Using an explicit command

When you specify a `command`, it completely replaces the default command:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment args-with-default-command}}
# Using args with default command
app_env = flyte.app.AppEnvironment(
    name="streamlit-app",
    args="streamlit run main.py --server.port 8080",
    port=8080,
    include=["main.py"],
    # command is None, so default Flyte command is used
)
# {{/docs-fragment args-with-default-command}}

# {{docs-fragment explicit-command}}
# Using explicit command
app_env2 = flyte.app.AppEnvironment(
    name="streamlit-hello",
    command="streamlit hello --server.port 8080",
    port=8080,
    # No args needed since command includes everything
)
# {{/docs-fragment explicit-command}}

# {{docs-fragment command-with-args}}
# Using command with args
app_env3 = flyte.app.AppEnvironment(
    name="custom-app",
    command="python -m myapp",
    args="--option1 value1 --option2 value2",
    # This runs: python -m myapp --option1 value1 --option2 value2
)
# {{/docs-fragment command-with-args}}

# {{docs-fragment fastapi-auto-command}}
# FastAPIAppEnvironment automatically sets command
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

app = FastAPI()

env = FastAPIAppEnvironment(
    name="my-api",
    app=app,
    # You typically don't need to specify command or args, since the
    # FastAPIAppEnvironment automatically uses the bundled code to serve the
    # app via uvicorn.
)
# {{/docs-fragment fastapi-auto-command}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/app-startup-examples.py*

This runs exactly:

```bash
streamlit hello --server.port 8080
```

#### Using a command with args

You can combine both, though this is less common:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment args-with-default-command}}
# Using args with default command
app_env = flyte.app.AppEnvironment(
    name="streamlit-app",
    args="streamlit run main.py --server.port 8080",
    port=8080,
    include=["main.py"],
    # command is None, so default Flyte command is used
)
# {{/docs-fragment args-with-default-command}}

# {{docs-fragment explicit-command}}
# Using explicit command
app_env2 = flyte.app.AppEnvironment(
    name="streamlit-hello",
    command="streamlit hello --server.port 8080",
    port=8080,
    # No args needed since command includes everything
)
# {{/docs-fragment explicit-command}}

# {{docs-fragment command-with-args}}
# Using command with args
app_env3 = flyte.app.AppEnvironment(
    name="custom-app",
    command="python -m myapp",
    args="--option1 value1 --option2 value2",
    # This runs: python -m myapp --option1 value1 --option2 value2
)
# {{/docs-fragment command-with-args}}

# {{docs-fragment fastapi-auto-command}}
# FastAPIAppEnvironment automatically sets command
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

app = FastAPI()

env = FastAPIAppEnvironment(
    name="my-api",
    app=app,
    # You typically don't need to specify command or args, since the
    # FastAPIAppEnvironment automatically uses the bundled code to serve the
    # app via uvicorn.
)
# {{/docs-fragment fastapi-auto-command}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/app-startup-examples.py*

#### FastAPIAppEnvironment example

When using `FastAPIAppEnvironment`, the command is automatically configured to run uvicorn:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment args-with-default-command}}
# Using args with default command
app_env = flyte.app.AppEnvironment(
    name="streamlit-app",
    args="streamlit run main.py --server.port 8080",
    port=8080,
    include=["main.py"],
    # command is None, so default Flyte command is used
)
# {{/docs-fragment args-with-default-command}}

# {{docs-fragment explicit-command}}
# Using explicit command
app_env2 = flyte.app.AppEnvironment(
    name="streamlit-hello",
    command="streamlit hello --server.port 8080",
    port=8080,
    # No args needed since command includes everything
)
# {{/docs-fragment explicit-command}}

# {{docs-fragment command-with-args}}
# Using command with args
app_env3 = flyte.app.AppEnvironment(
    name="custom-app",
    command="python -m myapp",
    args="--option1 value1 --option2 value2",
    # This runs: python -m myapp --option1 value1 --option2 value2
)
# {{/docs-fragment command-with-args}}

# {{docs-fragment fastapi-auto-command}}
# FastAPIAppEnvironment automatically sets command
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

app = FastAPI()

env = FastAPIAppEnvironment(
    name="my-api",
    app=app,
    # You typically don't need to specify command or args, since the
    # FastAPIAppEnvironment automatically uses the bundled code to serve the
    # app via uvicorn.
)
# {{/docs-fragment fastapi-auto-command}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/app-startup-examples.py*

The `FastAPIAppEnvironment` automatically:

1. Detects the module and variable name of your FastAPI app
2. Uses an internal server function to start the app via `uvicorn.run`.
3. Handles all the startup configuration for you

## Shared settings

For more details on shared settings like images, resources, and secrets, refer to the [task configuration](../task-configuration/_index) documentation.

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps/including-additional-files ===

# Including additional files

When your app needs additional files beyond the main script (like utility modules, configuration files, or data files), you can use the `include` parameter to specify which files to bundle with your app.

## How include works

The `include` parameter takes a list of file paths (relative to the directory containing your app definition). These files are bundled together and made available in the app container at runtime.

```python
include=["main.py", "utils.py", "config.yaml"]
```

## When to use include

Use `include` when:

- Your app spans multiple Python files (modules)
- You have configuration files that your app needs
- You have data files or templates your app uses
- You want to ensure specific files are available in the container

> [!NOTE]
> If you're using specialized app environments like `FastAPIAppEnvironment`, Flyte automatically detects and includes the necessary files, so you may not need to specify `include` explicitly.

## Examples

### Multi-file Streamlit app

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

"""A custom Streamlit app with multiple files."""

import pathlib
import flyte
import flyte.app

# {{docs-fragment app-env}}
image = flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages(
    "streamlit==1.41.1",
    "pandas==2.2.3",
    "numpy==2.2.3",
)

app_env = flyte.app.AppEnvironment(
    name="streamlit-multi-file-app",
    image=image,
    args="streamlit run main.py --server.port 8080",
    port=8080,
    include=["main.py", "utils.py"],  # Include your app files
    resources=flyte.Resources(cpu="1", memory="1Gi"),
    requires_auth=False,
)
# {{/docs-fragment app-env}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    flyte.init_from_config(root_dir=pathlib.Path(__file__).parent)
    app = flyte.deploy(app_env)
    print(f"Deployed app: {app[0].summary_repr()}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/streamlit/multi_file_streamlit.py*

In this example:
- `main.py` is your main Streamlit app file
- `utils.py` contains helper functions used by `main.py`
- Both files are included in the app bundle

### Multi-file FastAPI app

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
# ]
# ///

"""Multi-file FastAPI app example."""

from fastapi import FastAPI
from module import function  # Import from another file
import pathlib

import flyte
from flyte.app.extras import FastAPIAppEnvironment

# {{docs-fragment app-definition}}
app = FastAPI(title="Multi-file FastAPI Demo")

app_env = FastAPIAppEnvironment(
    name="fastapi-multi-file",
    app=app,
    image=flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages(
        "fastapi",
        "uvicorn",
    ),
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
    # FastAPIAppEnvironment automatically includes necessary files
    # But you can also specify explicitly:
    # include=["app.py", "module.py"],
)
# {{/docs-fragment app-definition}}

# {{docs-fragment endpoint}}
@app.get("/")
async def root():
    return function()  # Uses function from module.py
# {{/docs-fragment endpoint}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    flyte.init_from_config(root_dir=pathlib.Path(__file__).parent)
    app_deployment = flyte.deploy(app_env)
    print(f"Deployed: {app_deployment[0].summary_repr()}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/fastapi/multi_file/app.py*

### App with configuration files

```python
include=["app.py", "config.yaml", "templates/"]
```

## File discovery

When using specialized app environments like `FastAPIAppEnvironment`, Flyte uses code introspection to automatically discover and include the necessary files. This means you often don't need to manually specify `include`.

However, if you have files that aren't automatically detected (like configuration files, data files, or templates), you should explicitly list them in `include`.

## Path resolution

Files in `include` are resolved relative to the directory containing your app definition file. For example:

```
project/
├── apps/
│   ├── app.py          # Your app definition
│   ├── utils.py        # Included file
│   └── config.yaml     # Included file
```

In `app.py`:

```python
include=["utils.py", "config.yaml"]  # Relative to apps/ directory
```

## Best practices

1. **Only include what you need**: Don't include unnecessary files as it increases bundle size
2. **Use relative paths**: Always use paths relative to your app definition file
3. **Include directories**: You can include entire directories, but be mindful of size
4. **Test locally**: Verify your includes work by testing locally before deploying
5. **Check automatic discovery**: Specialized app environments may already include files automatically

## Limitations

- Large files or directories can slow down deployment
- Binary files are supported but consider using data storage (S3, etc.) for very large files
- The bundle size is limited by your Flyte cluster configuration

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps/passing-parameters ===

# Passing parameters into app environments

`[[AppEnvironment]]`s support various parameter types that can be passed at deployment time. This includes primitive values, files, directories, and delayed values like `RunOutput` and `AppEndpoint`.

## Parameter types overview

There are several parameter types:

- **Primitive values**: Strings, numbers, booleans
- **Files**: `flyte.io.File` objects
- **Directories**: `flyte.io.Dir` objects
- **Delayed values**: `RunOutput` (from task runs) or `AppEndpoint` (inject endpoint urls of other apps)

## Basic parameter types

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
#    "scikit-learn",
#    "joblib",
# ]
# ///

"""Examples showing different ways to pass parameters into apps."""

import flyte
import flyte.app
import flyte.io

# {{docs-fragment basic-parameter-types}}
# String parameters
app_env = flyte.app.AppEnvironment(
    name="configurable-app",
    parameters=[
        flyte.app.Parameter(name="environment", value="production"),
        flyte.app.Parameter(name="log_level", value="INFO"),
    ],
    # ...
)

# File parameters
app_env2 = flyte.app.AppEnvironment(
    name="app-with-model",
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            value=flyte.io.File("s3://bucket/models/model.pkl"),
            mount="/app/models",
        ),
    ],
    # ...
)

# Directory parameters
app_env3 = flyte.app.AppEnvironment(
    name="app-with-data",
    parameters=[
        flyte.app.Parameter(
            name="data_dir",
            value=flyte.io.Dir("s3://bucket/data/"),
            mount="/app/data",
        ),
    ],
    # ...
)
# {{/docs-fragment basic-parameter-types}}

# {{docs-fragment runoutput-example}}
# Delayed parameters with RunOutput
env = flyte.TaskEnvironment(name="training-env")

@env.task
async def train_model() -> flyte.io.File:
    # ... training logic ...
    return await flyte.io.File.from_local("/tmp/trained-model.pkl")

# Use the task output as an app parameter
app_env4 = flyte.app.AppEnvironment(
    name="serving-app",
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(type="file", run_name="training_run", task_name="train_model"),
            mount="/app/model",
        ),
    ],
    # ...
)
# {{/docs-fragment runoutput-example}}

# {{docs-fragment appendpoint-example}}
# Delayed parameters with AppEndpoint
app1_env = flyte.app.AppEnvironment(name="backend-api")

app2_env = flyte.app.AppEnvironment(
    name="frontend-app",
    parameters=[
        flyte.app.Parameter(
            name="backend_url",
            value=flyte.app.AppEndpoint(app_name="backend-api"),
            env_var="BACKEND_URL",  # app1_env's endpoint will be available as an environment variable
        ),
    ],
    # ...
)
# {{/docs-fragment appendpoint-example}}

# {{docs-fragment runoutput-serving-example}}
# Example: Using RunOutput for model serving
import joblib
from sklearn.ensemble import RandomForestClassifier
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

# Training task
training_env = flyte.TaskEnvironment(name="training-env")

@training_env.task
async def train_model_task() -> flyte.io.File:
    """Train a model and return it."""

    model = RandomForestClassifier()

    # ... training logic ...

    path = "./trained-model.pkl"
    joblib.dump(model, path)
    return await flyte.io.File.from_local(path)

# Serving app that uses the trained model
app = FastAPI()
serving_env = FastAPIAppEnvironment(
    name="model-serving-app",
    app=app,
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(
                type="file",
                task_name="training-env.train_model_task"
            ),
            mount="/app/model",
            env_var="MODEL_PATH",
        ),
    ],
)
# {{/docs-fragment runoutput-serving-example}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/passing-parameters-examples.py*

## Delayed values

Delayed values are parameters whose actual values are materialized at deployment time.

### RunOutput

Use `RunOutput` to pass outputs from task runs as app parameters:

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
#    "scikit-learn",
#    "joblib",
# ]
# ///

"""Examples showing different ways to pass parameters into apps."""

import flyte
import flyte.app
import flyte.io

# {{docs-fragment basic-parameter-types}}
# String parameters
app_env = flyte.app.AppEnvironment(
    name="configurable-app",
    parameters=[
        flyte.app.Parameter(name="environment", value="production"),
        flyte.app.Parameter(name="log_level", value="INFO"),
    ],
    # ...
)

# File parameters
app_env2 = flyte.app.AppEnvironment(
    name="app-with-model",
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            value=flyte.io.File("s3://bucket/models/model.pkl"),
            mount="/app/models",
        ),
    ],
    # ...
)

# Directory parameters
app_env3 = flyte.app.AppEnvironment(
    name="app-with-data",
    parameters=[
        flyte.app.Parameter(
            name="data_dir",
            value=flyte.io.Dir("s3://bucket/data/"),
            mount="/app/data",
        ),
    ],
    # ...
)
# {{/docs-fragment basic-parameter-types}}

# {{docs-fragment runoutput-example}}
# Delayed parameters with RunOutput
env = flyte.TaskEnvironment(name="training-env")

@env.task
async def train_model() -> flyte.io.File:
    # ... training logic ...
    return await flyte.io.File.from_local("/tmp/trained-model.pkl")

# Use the task output as an app parameter
app_env4 = flyte.app.AppEnvironment(
    name="serving-app",
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(type="file", run_name="training_run", task_name="train_model"),
            mount="/app/model",
        ),
    ],
    # ...
)
# {{/docs-fragment runoutput-example}}

# {{docs-fragment appendpoint-example}}
# Delayed parameters with AppEndpoint
app1_env = flyte.app.AppEnvironment(name="backend-api")

app2_env = flyte.app.AppEnvironment(
    name="frontend-app",
    parameters=[
        flyte.app.Parameter(
            name="backend_url",
            value=flyte.app.AppEndpoint(app_name="backend-api"),
            env_var="BACKEND_URL",  # app1_env's endpoint will be available as an environment variable
        ),
    ],
    # ...
)
# {{/docs-fragment appendpoint-example}}

# {{docs-fragment runoutput-serving-example}}
# Example: Using RunOutput for model serving
import joblib
from sklearn.ensemble import RandomForestClassifier
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

# Training task
training_env = flyte.TaskEnvironment(name="training-env")

@training_env.task
async def train_model_task() -> flyte.io.File:
    """Train a model and return it."""

    model = RandomForestClassifier()

    # ... training logic ...

    path = "./trained-model.pkl"
    joblib.dump(model, path)
    return await flyte.io.File.from_local(path)

# Serving app that uses the trained model
app = FastAPI()
serving_env = FastAPIAppEnvironment(
    name="model-serving-app",
    app=app,
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(
                type="file",
                task_name="training-env.train_model_task"
            ),
            mount="/app/model",
            env_var="MODEL_PATH",
        ),
    ],
)
# {{/docs-fragment runoutput-serving-example}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/passing-parameters-examples.py*

The `type` argument is required and must be one of `string`, `file`, or `directory`.
When the app is deployed, it will make the remote calls needed to figure out the
actual value of the parameter.

### AppEndpoint

Use `AppEndpoint` to pass endpoints from other apps:

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
#    "scikit-learn",
#    "joblib",
# ]
# ///

"""Examples showing different ways to pass parameters into apps."""

import flyte
import flyte.app
import flyte.io

# {{docs-fragment basic-parameter-types}}
# String parameters
app_env = flyte.app.AppEnvironment(
    name="configurable-app",
    parameters=[
        flyte.app.Parameter(name="environment", value="production"),
        flyte.app.Parameter(name="log_level", value="INFO"),
    ],
    # ...
)

# File parameters
app_env2 = flyte.app.AppEnvironment(
    name="app-with-model",
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            value=flyte.io.File("s3://bucket/models/model.pkl"),
            mount="/app/models",
        ),
    ],
    # ...
)

# Directory parameters
app_env3 = flyte.app.AppEnvironment(
    name="app-with-data",
    parameters=[
        flyte.app.Parameter(
            name="data_dir",
            value=flyte.io.Dir("s3://bucket/data/"),
            mount="/app/data",
        ),
    ],
    # ...
)
# {{/docs-fragment basic-parameter-types}}

# {{docs-fragment runoutput-example}}
# Delayed parameters with RunOutput
env = flyte.TaskEnvironment(name="training-env")

@env.task
async def train_model() -> flyte.io.File:
    # ... training logic ...
    return await flyte.io.File.from_local("/tmp/trained-model.pkl")

# Use the task output as an app parameter
app_env4 = flyte.app.AppEnvironment(
    name="serving-app",
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(type="file", run_name="training_run", task_name="train_model"),
            mount="/app/model",
        ),
    ],
    # ...
)
# {{/docs-fragment runoutput-example}}

# {{docs-fragment appendpoint-example}}
# Delayed parameters with AppEndpoint
app1_env = flyte.app.AppEnvironment(name="backend-api")

app2_env = flyte.app.AppEnvironment(
    name="frontend-app",
    parameters=[
        flyte.app.Parameter(
            name="backend_url",
            value=flyte.app.AppEndpoint(app_name="backend-api"),
            env_var="BACKEND_URL",  # app1_env's endpoint will be available as an environment variable
        ),
    ],
    # ...
)
# {{/docs-fragment appendpoint-example}}

# {{docs-fragment runoutput-serving-example}}
# Example: Using RunOutput for model serving
import joblib
from sklearn.ensemble import RandomForestClassifier
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

# Training task
training_env = flyte.TaskEnvironment(name="training-env")

@training_env.task
async def train_model_task() -> flyte.io.File:
    """Train a model and return it."""

    model = RandomForestClassifier()

    # ... training logic ...

    path = "./trained-model.pkl"
    joblib.dump(model, path)
    return await flyte.io.File.from_local(path)

# Serving app that uses the trained model
app = FastAPI()
serving_env = FastAPIAppEnvironment(
    name="model-serving-app",
    app=app,
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(
                type="file",
                task_name="training-env.train_model_task"
            ),
            mount="/app/model",
            env_var="MODEL_PATH",
        ),
    ],
)
# {{/docs-fragment runoutput-serving-example}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/passing-parameters-examples.py*

The endpoint URL will be injected as the parameter value when the app starts.

This is particularly useful when you want to chain apps together (for example, a frontend app calling a backend app), without hardcoding URLs.

## Overriding parameters at serve time

You can override parameter values when serving apps (this is not supported for deployment):

```python
# Override parameters when serving
app = flyte.with_servecontext(
    input_values={"my-app": {"model_path": "s3://bucket/new-model.pkl"}}
).serve(app_env)
```

> [!NOTE]
> Parameter overrides are only available when using `flyte.serve()` or `flyte.with_servecontext().serve()`. 
> The `flyte.deploy()` function does not support parameter overrides - parameters must be specified in the `AppEnvironment` definition.

This is useful for:
- Testing different configurations during development
- Using different models or data sources for testing
- A/B testing different app configurations

## Example: FastAPI app with configurable model

Here's a complete example showing how to use parameters in a FastAPI app:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "fastapi",
#    "uvicorn",
#    "joblib",
#    "scikit-learn",
#    "flyte>=2.0.0b52",
# ]
# ///

from contextlib import asynccontextmanager
from pathlib import Path

import flyte
import flyte.app
import flyte.io
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

# {{docs-fragment model-serving-api}}

image = flyte.Image.from_uv_script(__file__, name="app-parameters-fastapi-example")

task_env = flyte.TaskEnvironment(
    name="model_serving_task",
    image=image,
    resources=flyte.Resources(cpu=2, memory="1Gi"),
    cache="auto",
)

@task_env.task
async def train_model_task() -> flyte.io.File:
    """Train a model and return it."""
    import joblib
    import sklearn.ensemble
    import sklearn.datasets

    X, y = sklearn.datasets.make_classification(n_samples=1000, n_features=5, n_classes=2, random_state=42)
    model = sklearn.ensemble.RandomForestClassifier()
    model.fit(X, y)

    model_dir = Path("/tmp/model")
    model_dir.mkdir(parents=True, exist_ok=True)
    model_path = model_dir / "model.joblib"
    joblib.dump(model, model_path)
    return await flyte.io.File.from_local(model_path)

state = {}

@asynccontextmanager
async def lifespan(app: FastAPI):
    import joblib

    model = joblib.load("/root/models/model.joblib")
    state["model"] = model
    yield

app = FastAPI(lifespan=lifespan)

app_env = FastAPIAppEnvironment(
    name="model-serving-api",
    app=app,
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            # this is a placeholder
            value=flyte.io.File.from_existing_remote("s3://bucket/models/default.pkl"),
            mount="/root/models/",
            download=True,
        ),
    ],
    image=image,
    resources=flyte.Resources(cpu=2, memory="2Gi"),
    requires_auth=False,
)

@app.post("/predict")
async def predict(data: list[float]) -> dict[str, list[float]]:
    model = state["model"]
    return {"prediction": model.predict([data]).tolist()}

if __name__ == "__main__":
    import logging

    flyte.init_from_config(log_level=logging.DEBUG)

    run = flyte.run(train_model_task)
    print(f"Run: {run.url}")
    run.wait()

    model_file = run.outputs()[0]
    print(f"Model file: {model_file.path}")

    app = flyte.with_servecontext(
        parameter_values={
            "model-serving-api": {
                "model_file": flyte.io.File.from_existing_remote(model_file.path)
            }
        }
    ).serve(app_env)
    print(f"API URL: {app.url}")
# {{/docs-fragment model-serving-api}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/app-parameters-fastapi-example.py*

## Example: Using RunOutput for model serving

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
#    "scikit-learn",
#    "joblib",
# ]
# ///

"""Examples showing different ways to pass parameters into apps."""

import flyte
import flyte.app
import flyte.io

# {{docs-fragment basic-parameter-types}}
# String parameters
app_env = flyte.app.AppEnvironment(
    name="configurable-app",
    parameters=[
        flyte.app.Parameter(name="environment", value="production"),
        flyte.app.Parameter(name="log_level", value="INFO"),
    ],
    # ...
)

# File parameters
app_env2 = flyte.app.AppEnvironment(
    name="app-with-model",
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            value=flyte.io.File("s3://bucket/models/model.pkl"),
            mount="/app/models",
        ),
    ],
    # ...
)

# Directory parameters
app_env3 = flyte.app.AppEnvironment(
    name="app-with-data",
    parameters=[
        flyte.app.Parameter(
            name="data_dir",
            value=flyte.io.Dir("s3://bucket/data/"),
            mount="/app/data",
        ),
    ],
    # ...
)
# {{/docs-fragment basic-parameter-types}}

# {{docs-fragment runoutput-example}}
# Delayed parameters with RunOutput
env = flyte.TaskEnvironment(name="training-env")

@env.task
async def train_model() -> flyte.io.File:
    # ... training logic ...
    return await flyte.io.File.from_local("/tmp/trained-model.pkl")

# Use the task output as an app parameter
app_env4 = flyte.app.AppEnvironment(
    name="serving-app",
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(type="file", run_name="training_run", task_name="train_model"),
            mount="/app/model",
        ),
    ],
    # ...
)
# {{/docs-fragment runoutput-example}}

# {{docs-fragment appendpoint-example}}
# Delayed parameters with AppEndpoint
app1_env = flyte.app.AppEnvironment(name="backend-api")

app2_env = flyte.app.AppEnvironment(
    name="frontend-app",
    parameters=[
        flyte.app.Parameter(
            name="backend_url",
            value=flyte.app.AppEndpoint(app_name="backend-api"),
            env_var="BACKEND_URL",  # app1_env's endpoint will be available as an environment variable
        ),
    ],
    # ...
)
# {{/docs-fragment appendpoint-example}}

# {{docs-fragment runoutput-serving-example}}
# Example: Using RunOutput for model serving
import joblib
from sklearn.ensemble import RandomForestClassifier
from flyte.app.extras import FastAPIAppEnvironment
from fastapi import FastAPI

# Training task
training_env = flyte.TaskEnvironment(name="training-env")

@training_env.task
async def train_model_task() -> flyte.io.File:
    """Train a model and return it."""

    model = RandomForestClassifier()

    # ... training logic ...

    path = "./trained-model.pkl"
    joblib.dump(model, path)
    return await flyte.io.File.from_local(path)

# Serving app that uses the trained model
app = FastAPI()
serving_env = FastAPIAppEnvironment(
    name="model-serving-app",
    app=app,
    parameters=[
        flyte.app.Parameter(
            name="model",
            value=flyte.app.RunOutput(
                type="file",
                task_name="training-env.train_model_task"
            ),
            mount="/app/model",
            env_var="MODEL_PATH",
        ),
    ],
)
# {{/docs-fragment runoutput-serving-example}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/passing-parameters-examples.py*

## Accessing parameters in your app

How you access parameters depends on how they're configured:

1. **Environment variables**: If `env_var` is specified, the parameter is available as an environment variable
2. **Mounted paths**: File and directory parameters are mounted at the specified path
3. **Flyte SDK**: Use the Flyte SDK to access parameter values programmatically

```python
import os

# Parameter with env_var specified
env = flyte.app.AppEnvironment(
    name="my-app",
    parameters=[
        flyte.app.Parameter(
            name="model_file",
            value=flyte.io.File("s3://bucket/model.pkl"),
            mount="/app/models/model.pkl",
            env_var="MODEL_PATH",
        ),
    ],
    # ...
)

# Access in the app via the environment variable
API_KEY = os.getenv("API_KEY")

# Access in the app via the mounted path
with open("/app/models/model.pkl", "rb") as f:
    model = pickle.load(f)

# Access in the app via the Flyte SDK (for string parameters)
parameter_value = flyte.app.get_parameter("model_file")  # Returns string value
```

## Best practices

1. **Use delayed parameters**: Leverage `RunOutput` and `AppEndpoint` to create app dependencies between tasks and apps, or app-to-app chains.
2. **Override for testing**: Use the `input_values` parameter when serving to test different configurations without changing code.
3. **Mount paths clearly**: Use descriptive mount paths for file/directory parameters so your app code is easy to understand.
4. **Use environment variables**: For simple constants that you can hard-code, use `env_var` to inject values as environment variables.
5. **Production deployments**: For production, define parameters in the `AppEnvironment` rather than overriding them at deploy time.

## Limitations

- Large files/directories can slow down app startup.
- Parameter overrides are only available when using `flyte.with_servecontext(...).serve(...)`.

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps/auto-scaling-apps ===

## Autoscaling apps

Flyte apps support autoscaling, allowing them to scale up and down based on traffic. This helps optimize costs by scaling down when there's no traffic and scaling up when needed.

### Scaling configuration

The `scaling` parameter uses a `[[Scaling]]` object to configure autoscaling behavior:

```python
scaling=flyte.app.Scaling(
    replicas=(min_replicas, max_replicas),
    scaledown_after=idle_ttl_seconds,
)
```

#### Parameters

- **`replicas`**: A tuple `(min_replicas, max_replicas)` specifying the minimum and maximum number of replicas.
- **`scaledown_after`**: Time in seconds to wait before scaling down when idle (idle TTL).

### Basic scaling example

Here's a simple example with scaling from 0 to 1 replica:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment basic-scaling}}
# Basic example: scale from 0 to 1 replica
app_env = flyte.app.AppEnvironment(
    name="autoscaling-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Scale from 0 to 1 replica
        scaledown_after=300,  # Scale down after 5 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment basic-scaling}}

# {{docs-fragment always-on}}
# Always-on app
app_env2 = flyte.app.AppEnvironment(
    name="always-on-api",
    scaling=flyte.app.Scaling(
        replicas=(1, 1),  # Always keep 1 replica running
        # scaledown_after is ignored when min_replicas > 0
    ),
    # ...
)
# {{/docs-fragment always-on}}

# {{docs-fragment scale-to-zero}}
# Scale-to-zero app
app_env3 = flyte.app.AppEnvironment(
    name="scale-to-zero-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Can scale down to 0
        scaledown_after=600,  # Scale down after 10 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment scale-to-zero}}

# {{docs-fragment high-availability}}
# High-availability app
app_env4 = flyte.app.AppEnvironment(
    name="ha-api",
    scaling=flyte.app.Scaling(
        replicas=(2, 5),  # Keep at least 2, scale up to 5
        scaledown_after=300,  # Scale down after 5 minutes
    ),
    # ...
)
# {{/docs-fragment high-availability}}

# {{docs-fragment burstable}}
# Burstable app
app_env5 = flyte.app.AppEnvironment(
    name="bursty-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 10),  # Start with 1, scale up to 10 under load
        scaledown_after=180,  # Scale down quickly after 3 minutes
    ),
    # ...
)
# {{/docs-fragment burstable}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/autoscaling-examples.py*

This configuration:

- Starts with 0 replicas (no running instances)
- Scales up to 1 replica when there's traffic
- Scales back down to 0 after 5 minutes (300 seconds) of no traffic

### Scaling patterns

#### Always-on app

For apps that need to always be running:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment basic-scaling}}
# Basic example: scale from 0 to 1 replica
app_env = flyte.app.AppEnvironment(
    name="autoscaling-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Scale from 0 to 1 replica
        scaledown_after=300,  # Scale down after 5 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment basic-scaling}}

# {{docs-fragment always-on}}
# Always-on app
app_env2 = flyte.app.AppEnvironment(
    name="always-on-api",
    scaling=flyte.app.Scaling(
        replicas=(1, 1),  # Always keep 1 replica running
        # scaledown_after is ignored when min_replicas > 0
    ),
    # ...
)
# {{/docs-fragment always-on}}

# {{docs-fragment scale-to-zero}}
# Scale-to-zero app
app_env3 = flyte.app.AppEnvironment(
    name="scale-to-zero-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Can scale down to 0
        scaledown_after=600,  # Scale down after 10 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment scale-to-zero}}

# {{docs-fragment high-availability}}
# High-availability app
app_env4 = flyte.app.AppEnvironment(
    name="ha-api",
    scaling=flyte.app.Scaling(
        replicas=(2, 5),  # Keep at least 2, scale up to 5
        scaledown_after=300,  # Scale down after 5 minutes
    ),
    # ...
)
# {{/docs-fragment high-availability}}

# {{docs-fragment burstable}}
# Burstable app
app_env5 = flyte.app.AppEnvironment(
    name="bursty-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 10),  # Start with 1, scale up to 10 under load
        scaledown_after=180,  # Scale down quickly after 3 minutes
    ),
    # ...
)
# {{/docs-fragment burstable}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/autoscaling-examples.py*

#### Scale-to-zero app

For apps that can scale to zero when idle:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment basic-scaling}}
# Basic example: scale from 0 to 1 replica
app_env = flyte.app.AppEnvironment(
    name="autoscaling-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Scale from 0 to 1 replica
        scaledown_after=300,  # Scale down after 5 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment basic-scaling}}

# {{docs-fragment always-on}}
# Always-on app
app_env2 = flyte.app.AppEnvironment(
    name="always-on-api",
    scaling=flyte.app.Scaling(
        replicas=(1, 1),  # Always keep 1 replica running
        # scaledown_after is ignored when min_replicas > 0
    ),
    # ...
)
# {{/docs-fragment always-on}}

# {{docs-fragment scale-to-zero}}
# Scale-to-zero app
app_env3 = flyte.app.AppEnvironment(
    name="scale-to-zero-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Can scale down to 0
        scaledown_after=600,  # Scale down after 10 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment scale-to-zero}}

# {{docs-fragment high-availability}}
# High-availability app
app_env4 = flyte.app.AppEnvironment(
    name="ha-api",
    scaling=flyte.app.Scaling(
        replicas=(2, 5),  # Keep at least 2, scale up to 5
        scaledown_after=300,  # Scale down after 5 minutes
    ),
    # ...
)
# {{/docs-fragment high-availability}}

# {{docs-fragment burstable}}
# Burstable app
app_env5 = flyte.app.AppEnvironment(
    name="bursty-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 10),  # Start with 1, scale up to 10 under load
        scaledown_after=180,  # Scale down quickly after 3 minutes
    ),
    # ...
)
# {{/docs-fragment burstable}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/autoscaling-examples.py*

#### High-availability app

For apps that need multiple replicas for availability:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment basic-scaling}}
# Basic example: scale from 0 to 1 replica
app_env = flyte.app.AppEnvironment(
    name="autoscaling-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Scale from 0 to 1 replica
        scaledown_after=300,  # Scale down after 5 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment basic-scaling}}

# {{docs-fragment always-on}}
# Always-on app
app_env2 = flyte.app.AppEnvironment(
    name="always-on-api",
    scaling=flyte.app.Scaling(
        replicas=(1, 1),  # Always keep 1 replica running
        # scaledown_after is ignored when min_replicas > 0
    ),
    # ...
)
# {{/docs-fragment always-on}}

# {{docs-fragment scale-to-zero}}
# Scale-to-zero app
app_env3 = flyte.app.AppEnvironment(
    name="scale-to-zero-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Can scale down to 0
        scaledown_after=600,  # Scale down after 10 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment scale-to-zero}}

# {{docs-fragment high-availability}}
# High-availability app
app_env4 = flyte.app.AppEnvironment(
    name="ha-api",
    scaling=flyte.app.Scaling(
        replicas=(2, 5),  # Keep at least 2, scale up to 5
        scaledown_after=300,  # Scale down after 5 minutes
    ),
    # ...
)
# {{/docs-fragment high-availability}}

# {{docs-fragment burstable}}
# Burstable app
app_env5 = flyte.app.AppEnvironment(
    name="bursty-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 10),  # Start with 1, scale up to 10 under load
        scaledown_after=180,  # Scale down quickly after 3 minutes
    ),
    # ...
)
# {{/docs-fragment burstable}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/autoscaling-examples.py*

#### Burstable app

For apps with variable load:

```
# /// script
# requires-python = "==3.13"
# dependencies = [
#    "flyte>=2.0.0b52",
# ]
# ///

import flyte
import flyte.app

# {{docs-fragment basic-scaling}}
# Basic example: scale from 0 to 1 replica
app_env = flyte.app.AppEnvironment(
    name="autoscaling-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Scale from 0 to 1 replica
        scaledown_after=300,  # Scale down after 5 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment basic-scaling}}

# {{docs-fragment always-on}}
# Always-on app
app_env2 = flyte.app.AppEnvironment(
    name="always-on-api",
    scaling=flyte.app.Scaling(
        replicas=(1, 1),  # Always keep 1 replica running
        # scaledown_after is ignored when min_replicas > 0
    ),
    # ...
)
# {{/docs-fragment always-on}}

# {{docs-fragment scale-to-zero}}
# Scale-to-zero app
app_env3 = flyte.app.AppEnvironment(
    name="scale-to-zero-app",
    scaling=flyte.app.Scaling(
        replicas=(0, 1),  # Can scale down to 0
        scaledown_after=600,  # Scale down after 10 minutes of inactivity
    ),
    # ...
)
# {{/docs-fragment scale-to-zero}}

# {{docs-fragment high-availability}}
# High-availability app
app_env4 = flyte.app.AppEnvironment(
    name="ha-api",
    scaling=flyte.app.Scaling(
        replicas=(2, 5),  # Keep at least 2, scale up to 5
        scaledown_after=300,  # Scale down after 5 minutes
    ),
    # ...
)
# {{/docs-fragment high-availability}}

# {{docs-fragment burstable}}
# Burstable app
app_env5 = flyte.app.AppEnvironment(
    name="bursty-app",
    scaling=flyte.app.Scaling(
        replicas=(1, 10),  # Start with 1, scale up to 10 under load
        scaledown_after=180,  # Scale down quickly after 3 minutes
    ),
    # ...
)
# {{/docs-fragment burstable}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/configure-apps/autoscaling-examples.py*

### Idle TTL (Time To Live)

The `scaledown_after` parameter (idle TTL) determines how long an app instance can be idle before it's scaled down. 

#### Considerations

- **Too short**: May cause frequent scale up/down cycles, leading to cold starts.
- **Too long**: Keeps resources running unnecessarily, increasing costs.
- **Optimal**: Balance between cost and user experience.

#### Common idle TTL values

- **Development/Testing**: 60-180 seconds (1-3 minutes) - quick scale down for cost savings.
- **Production APIs**: 300-600 seconds (5-10 minutes) - balance cost and responsiveness.
- **Batch processing**: 900-1800 seconds (15-30 minutes) - longer to handle bursts.
- **Always-on**: Set `min_replicas > 0` - never scale down.

### Autoscaling best practices

1. **Start conservative**: Begin with longer idle TTL values and adjust based on usage.
2. **Monitor cold starts**: Track how long it takes for your app to become ready after scaling up.
3. **Consider costs**: Balance idle TTL between cost savings and user experience.
4. **Use appropriate min replicas**: Set `min_replicas > 0` for critical apps that need to be always available.
5. **Test scaling behavior**: Verify your app handles scale up/down correctly (for example, state management and connections).

### Autoscaling limitations

- Scaling is based on traffic/request patterns, not CPU/memory utilization.
- Cold starts may occur when scaling from zero.
- Stateful apps need careful design to handle scaling (use external state stores).
- Maximum replicas are limited by your cluster capacity.

### Autoscaling troubleshooting

**App scales down too quickly:**

- Increase `scaledown_after` value.
- Set `min_replicas > 0` if the app needs to stay warm.

**App doesn't scale up fast enough:**

- Ensure your cluster has capacity.
- Check if there are resource constraints.

**Cold starts are too slow:**

- Pre-warm with `min_replicas = 1`.
- Optimize app startup time.
- Consider using faster storage for model loading.

=== PAGE: https://www.union.ai/docs/v2/union/user-guide/configure-apps/apps-depending-on-environments ===

# Apps depending on other environments

The `depends_on` parameter allows you to specify that one app depends on another app (or task environment). When you deploy an app with `depends_on`, Flyte ensures that all dependencies are deployed first.

## Basic usage

Use `depends_on` to specify a list of environments that this app depends on:

```python
app1_env = flyte.app.AppEnvironment(name="backend-api", ...)

app2_env = flyte.app.AppEnvironment(
    name="frontend-app",
    depends_on=[app1_env],  # Ensure backend-api is deployed first
    # ...
)
```

When you deploy `app2_env`, Flyte will:
1. First deploy `app1_env` (if not already deployed)
2. Then deploy `app2_env`
3. Make sure `app1_env` is available before `app2_env` starts

## Example: App calling another app

Here's a complete example where one FastAPI app calls another:

```
# /// script
# requires-python = ">=3.12"
# dependencies = [
#    "flyte>=2.0.0b52",
#    "fastapi",
#    "httpx",
# ]
# ///

"""Example of one app calling another app."""

import httpx
from fastapi import FastAPI
import pathlib
import flyte
from flyte.app.extras import FastAPIAppEnvironment

image = flyte.Image.from_debian_base(python_version=(3, 12)).with_pip_packages(
    "fastapi", "uvicorn", "httpx"
)

# {{docs-fragment backend-app}}
app1 = FastAPI(
    title="App 1",
    description="A FastAPI app that runs some computations",
)

env1 = FastAPIAppEnvironment(
    name="app1-is-called-by-app2",
    app=app1,
    image=image,
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
)
# {{/docs-fragment backend-app}}

# {{docs-fragment frontend-app}}
app2 = FastAPI(
    title="App 2",
    description="A FastAPI app that proxies requests to another FastAPI app",
)

env2 = FastAPIAppEnvironment(
    name="app2-calls-app1",
    app=app2,
    image=image,
    resources=flyte.Resources(cpu=1, memory="512Mi"),
    requires_auth=False,
    depends_on=[env1],  # Depends on backend-api
)
# {{/docs-fragment frontend-app}}

# {{docs-fragment backend-endpoint}}
@app1.get("/greeting/{name}")
async def greeting(name: str) -> str:
    return f"Hello, {name}!"
# {{/docs-fragment backend-endpoint}}

# {{docs-fragment frontend-endpoints}}
@app2.get("/app1-endpoint")
async def get_app1_endpoint() -> str:
    return env1.endpoint  # Access the backend endpoint

@app2.get("/greeting/{name}")
async def greeting_proxy(name: str):
    """Proxy that calls the backend app."""
    async with httpx.AsyncClient() as client:
        response = await client.get(f"{env1.endpoint}/greeting/{name}")
        response.raise_for_status()
        return response.json()
# {{/docs-fragment frontend-endpoints}}

# {{docs-fragment deploy}}
if __name__ == "__main__":
    flyte.init_from_config(root_dir=pathlib.Path(__file__).parent)
    deployments = flyte.deploy(env2)
    print(f"Deployed FastAPI app: {deployments[0].env_repr()}")
# {{/docs-fragment deploy}}
```

*Source: https://github.com/unionai/unionai-examples/blob/main/v2/user-guide/build-apps/fastapi/app_calling_app.py*

When you deploy `env2`, Flyte will:
1. Deploy `env1` first (backend-api)
2. Wait for `env1` to be ready
3. Deploy `env2` (frontend-api)
4. `env2` can then access `env1.endpoint` to make requests

## Dependency chain

You can create chains of dependencies:

```python
app1_env = flyte.app.AppEnvironment(name="service-1", ...)
app2_env = flyte.app.AppEnvironment(name="service-2", depends_on=[app1_env], ...)
app3_env = flyte.app.AppEnvironment(name="service-3", depends_on=[app2_env], ...)

# Deploying app3_env will deploy in order: app1_env -> app2_env -> app3_env
```

## Multiple dependencies

An app can depend on multiple environments:

```python
backend_env = flyte.app.AppEnvironment(name="backend", ...)
database_env = flyte.app.AppEnvironment(name="database", ...)

api_env = flyte.app.AppEnvironment(
    name="api",
    depends_on=[backend_env, database_env],  # Depends on both
    # ...
)
```

When deploying `api_env`, both `backend_env` and `database_env` will be deployed first (they may be deployed in parallel if they don't depend on each other).

## Using AppEndpoint for dependency URLs

When one app depends on another, you can use `AppEndpoint` to get the URL:

```python
backend_env = flyte.app.AppEnvironment(name="backend-api", ...)

frontend_env = flyte.app.AppEnvironment(
    name="frontend-app",
    depends_on=[backend_env],
    parameters=[
        flyte.app.Parameter(
            name="backend_url",
            value=flyte.app.AppEndpoint(app_name="backend-api"),
        ),
    ],
    # ...
)
```

The `backend_url` parameter will be automatically set to the backend app's endpoint URL.
You can get this value in your app code using `flyte.app.get_input("backend_url")`.

## Deployment behavior

When deploying with `flyte.deploy()`:

```python
# Deploy the app (dependencies are automatically deployed)
deployments = flyte.deploy(env2)

# All dependencies are included in the deployment plan
for deployment in deployments:
    print(f"Deployed: {deployment.env.name}")
```

Flyte will:
1. Build a deployment plan that includes all dependencies
2. Deploy dependencies in the correct order
3. Ensure dependencies are ready before deploying dependent apps

## Task environment dependencies

You can also depend on task environments:

```python
task_env = flyte.TaskEnvironment(name="training-env", ...)

serving_env = flyte.app.AppEnvironment(
    name="serving-app",
    depends_on=[task_env],  # Can depend on task environments too
    # ...
)
```

This ensures the task environment is available when the app is deployed (useful if the app needs to call tasks in that environment).

## Best practices

1. **Explicit dependencies**: Always use `depends_on` to make app dependencies explicit
2. **Circular dependencies**: Avoid circular dependencies (app A depends on B, B depends on A)
3. **Dependency order**: Design your dependency graph to be a DAG (Directed Acyclic Graph)
4. **Endpoint access**: Use `AppEndpoint` to pass dependency URLs as inputs
5. **Document dependencies**: Make sure your app documentation explains its dependencies

## Example: A/B testing with dependencies

Here's an example of an A/B testing setup where a root app depends on two variant apps:

```python
app_a = FastAPI(title="Variant A")
app_b = FastAPI(title="Variant B")
root_app = FastAPI(title="Root App")

env_a = FastAPIAppEnvironment(name="app-a-variant", app=app_a, ...)
env_b = FastAPIAppEnvironment(name="app-b-variant", app=app_b, ...)

env_root = FastAPIAppEnvironment(
    name="root-ab-testing-app",
    app=root_app,
    depends_on=[env_a, env_b],  # Depends on both variants
    # ...
)
```

The root app can route traffic to either variant A or B based on A/B testing logic, and both variants will be deployed before the root app starts.

## Limitations

- Circular dependencies are not supported
- Dependencies must be in the same project/domain
- Dependency deployment order is deterministic but dependencies at the same level may deploy in parallel