> ## Documentation Index
> Fetch the complete documentation index at: https://docs.knouds.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart: build and call your first Knouds workflow

> Sign up, build your first AI pipeline on the canvas, generate an API key, and make your first Knouds workflow API call in minutes.

This guide walks you through creating a Knouds account, building a simple image-generation pipeline on the canvas, and calling it via the REST API. By the end you'll have a working endpoint you can integrate into any application.

<Note>
  New accounts require admin approval before you can run workflows. After signing up and verifying your email, your account enters a pending state. You'll receive an email when your account is approved and ready to use.
</Note>

<Steps>
  <Step title="Sign up at knouds.ai">
    Go to [knouds.ai](https://knouds.ai) and create an account. You can sign up with your email address or with Google.

    After signing up, check your inbox for a verification email and click the confirmation link. Once your email is verified, your account will be reviewed for approval. You'll get a follow-up email when you're approved and can start building.
  </Step>

  <Step title="Build your first workflow">
    Once approved, open the canvas from your dashboard and create a new workflow. Give it a descriptive name — the name becomes part of your API endpoint URL and cannot be changed later.

    Add three nodes to the canvas:

    1. **Request Input node** — this defines the inputs your API accepts. Add a text field named `prompt`.
    2. **Image Generator node** — this runs the image model. Connect the `prompt` output from the Request Input node to the `prompt` input on the Image Generator.
    3. **Response Output node** — this collects the result and returns it in the API response. Connect the image output from the Image Generator to the Response Output node.

    Your canvas should show a left-to-right flow: Request Input → Image Generator → Response Output. Click **Run** to test the workflow on the canvas before deploying.
  </Step>

  <Step title="Deploy the workflow">
    Save your workflow using the toolbar. Knouds automatically makes the workflow available at:

    ```
    POST https://knouds.ai/api/workflows/YOUR-WORKFLOW-SLUG/run
    ```

    The slug is derived from the name you gave the workflow at creation time. You can find the exact slug in the workflow's URL when it's open in the editor.
  </Step>

  <Step title="Generate an API key">
    Open the [Developer Dashboard](https://knouds.ai/app/api/keys) (or **Settings → API Keys**, which redirects). Click **Create Key**, pick the **Workflow Deploy** preset, and copy the plaintext value — it's shown only once.

    Your tier determines what capabilities you can grant on a key:

    * **Free** — no external API access (canvas only). Upgrade to Pro to create usable keys.
    * **Pro** — `workflow:run`, `workflow:read`, `workflow:write`, `workflow:<slug>:run`, `webhook:receive`. 100 req/min.
    * **Business / Enterprise** — adds `model:run`, `model:<id>:run`, `agent:invoke`. 1,000 req/min (unlimited on Enterprise).

    The **Workflow Deploy** preset grants `workflow:run` + `workflow:read` — exactly what the next step needs.
  </Step>

  <Step title="Call your workflow via the API">
    Use `curl` to run your workflow. Replace `YOUR-SLUG` with your workflow's slug and `YOUR_API_KEY` with the key you just copied:

    ```bash theme={null}
    curl -X POST https://knouds.ai/api/workflows/YOUR-SLUG/run \
      -H "x-api-key: YOUR_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{"inputs": {"prompt": "a serene mountain landscape"}}'
    ```

    A successful response looks like this:

    ```json theme={null}
    {
      "success": true,
      "result": {
        "images": [
          { "url": "https://..." }
        ]
      },
      "cost": 0.003,
      "ms": 4210
    }
    ```

    * `result` contains the model output, including any media URLs
    * `cost` is the number of credits deducted from your balance for this run
    * `ms` is the total execution time in milliseconds
  </Step>
</Steps>

## Next steps

* Read the [authentication guide](/authentication) to understand capabilities and rate limits in detail.
* Explore [node types](/concepts/nodes) to see what else you can add to your pipelines.
* Check the [API reference](/api/overview) for all available endpoints, request shapes, and error codes.