# Installing the SDK

Workflow servers are the way the Ebbot platform can be extended by exposing actions that the Ebbot platform can take, and triggers that activate the Ebbot platform when certain events happen.

## Installation

We recommend that you use [uv](https://docs.astral.sh/uv/getting-started/installation/) when developing with the SDK. uv is a package manager and virtualenv built into one.

Start by creating a new project:

```bash
uv init workflow_server_demo
```

Then install the integrations\_sdk:

```
uv add integrations_sdk git+https://github.com/ebbot-ai/integrations_sdk.git@main
```

## Create the server

Add the following to a Python file, for example in the root of your project:

```python
from pydantic import BaseModel
from integrations_sdk.server import start_workflow_server


class Options(BaseModel):
    notSecret: str


class Secrets(BaseModel):
    secret: str

app = start_workflow_server(
    "fns",
    "http://localhost:9000",
    "engine-api-key",
    Options,
    Secrets,
    install_instructions="Docs on point, README magic",
    docs="My docs",
    auth_token="optional-shared-token",
)
```

Arguments:

* `module_name`: module path containing workflow actions and triggers (e.g. `"fns"`). This should be the path of a Python module that will be scanned for triggers and actions.
* `engine_base_url`: base URL for the Ebbot workflow engine. You can set this to any URL when testing your server.
* `engine_api_key`: bearer token used when the server calls the Ebbot workflow engine. You can set this to any value when testing.
* `Options`/`Secrets`: Pydantic models describing required connection data.
* `validator`: optional function to validate the `Options` and `Secrets` payload.
* `install_instructions`: This is shown when installing the workflow server. Markdown is allowed.
* `docs`: Provide documentation for this workflow server. This will be added as a documentation page on the ebbot platform docs. Markdown is allowed.
* `auth_token`: optional server auth; if set, requests must include `Authorization: Bearer <token>`.
* `dev_mode`: The Ebbot platform takes care of storing options and secrets for you in production. That is not available when testing the server locally, so we provide a dev\_mode that stores your information in a local SQLite database. Set this to true when testing your server.
* `post_install_instructions`: This is a callback that you can specify in order to provide post install instructions after a connection is created. It will be displayed in the UI after a connection has been established.

## Setting up the server for development

In this example, actions and triggers that you create will be located in the `fns` folder. Change this to your liking.

When developing and the `engine_base_url` and `engine_api_key` do not matter, you should also set `dev_mode=True` to allow you to store options and secrets locally.

```python
from pydantic import BaseModel
from integrations_sdk.server import start_workflow_server


class Options(BaseModel):
    notSecret: str


class Secrets(BaseModel):
    secret: str

app = start_workflow_server(
    "fns",
    "http://localhost:9000", # can be anything
    "engine-api-key", # can be anything
    Options,
    Secrets,
    dev_mode=True # Set to true.
)
```

### Starting the server

You can start the server locally by running the Python file you created with FastAPI. For example, if you named the file `main.py`:

```bash
[uv run] fastapi dev endeavour/main.py
```

### What the server exposes

The workflow server exposes a REST API that the Ebbot platform uses to interact with it. You can check the API by navigating to <http://localhost:8000/docs>

### Setting up your server for production

When hosting your server with Ebbot you need to provide a correct engine URL and API key. If we take care of hosting, we will provide those for your server through environment variables. Starting a production server should look like:

```python
from pydantic import BaseModel
from integrations_sdk.server import start_workflow_server
from os import getenv

class Options(BaseModel):
    notSecret: str


class Secrets(BaseModel):
    secret: str

app = start_workflow_server(
    "fns",
    getenv("CONNECTION_SERVER_URL"),
    getenv("CONNECTION_SERVER_KEY"),
    Options,
    Secrets,
    dev_mode=False
)
```

Congratulations on your good work so far. Start learning about how to develop locally through the link below.&#x20;

{% content-ref url="/pages/Ve43fLaHbzSQ6XSEerHM" %}
[Developing](/ebbot-docs/developer-resources/ebbot-automations/integrations-sdk/developing.md)
{% endcontent-ref %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.ebbot.ai/ebbot-docs/developer-resources/ebbot-automations/integrations-sdk/installing-the-sdk.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.