Prerequisites

Before you start, make sure you have the following at hand:

• An active amber instance with API access

• Your amber API key (found in the user settings under Access Keys)

• A running n8n instance (cloud or self-hosted)

Step 1: Retrieve API Key in amber

Open amber and navigate to User Settings → Access Keys. There you can generate your personal API key. Copy it — you’ll need it in n8n.

Note: Treat your API key like a password. Do not share it publicly and store it securely, as you can only view it once.

Step 2: Set up an HTTP Request node in n8n

amber offers a REST API that you can use to perform searches and fetch AI answers. In n8n, you’ll use the HTTP Request node for this.

1. Add a new node to your workflow and choose HTTP Request.

2. Set the method to POST.

3. Enter your amber API endpoint as the URL, for example:https://app.ambersearch.de/api/v1/chat

4. Navigate to Authentication and choose “Preferred Credential Type”.

5. A new field called “Credential Type” will appear where you should select “Bearer Auth”.

6. Click the field below and create a “New Credential”.

7. In “Bearer Auth,” paste your API key and click “Save”.

8. Under Headers, set the header Content-Type to application/json.

Step 3: Configure the request body

In the Body section of the HTTP Request node, choose the type JSON and enter your request. A simple example looks like this:

Example
HTML
Copy
   {
        "model": "telekom-gpt-4o",
        "temperature": 0.7,
        "stream": False,
        "tools": [],
        "messages": [
            {"role": "system", "content": "You are a helpful assistant."},
            {
                "role": "user",
                "content": "What would be the best activity to do in summer in Paris?",
            },
        ],
    },

You can populate the values dynamically by using n8n expressions, for example {{ $json.userInput }}, to take values from a previous node.

Step 4: Process the response

After a successful API call, amber returns a JSON response. You’ll find the AI result in the field answer. You can reuse this value in subsequent nodes — for example, to send it via Slack, write it into a table, or return it in an email.

An example of the output:

Example
HTML
Copy
{
  "content": "Laut unserer internen Richtlinie stehen dir 30 Urlaubstage pro Jahr zu.",
  "sources": [...]
}

Note: If you need the information in a specific format, you should add an additional Code node afterwards. This will help you obtain a “clean” JSON. Here is an example of such code with JavaScript:

Example
HTML
Copy
// Extract the assistant message content from AmberSearch response
const choices = $input.item.json.choices;
if (!choices || choices.length === 0) {
  throw new Error('No choices returned from AmberSearch API');
}

// The content may be a string or an object depending on the model
const rawContent = choices[0]?.message?.content ?? choices[0]?.text ?? '';

// Strip any accidental markdown fences
const cleaned = rawContent.replace(/```json|```/gi, '').trim();

let parsed;
try {
  parsed = JSON.parse(cleaned);
} catch (e) {
  // Fallback: return nulls rather than crashing the entire run
  parsed = { knowledge_workers: null, industry: null, annual_revenue_eur: null };
}

return {
  json: {
    'Contact ID': $('Loop Over Items1').item.json.id,
    'Company': parsed.company ?? null,
    'Location': parsed.company_location ?? null,
    'Knowledge Workers': parsed.knowledge_workers ?? null,
    'Industry': parsed.industry ?? null,
    'Sub-industry': parsed.sub_industry ?? null,
    'Annual Revenue (€)': parsed.annual_revenue_eur ?? null
  }
};

Example Workflow: Connect HubSpot with amber

A common use case is CRM management and data enrichment through a workflow:

• Trigger: A new HubSpot contact is created (HubSpot Trigger Node)

• Processing: HTTP request to amber with the message as the query

• Response: The automatically researched information from the amber response is stored in the HubSpot contact

This workflow can be built in n8n within a few minutes and enables your team to use amber’s AI functionality in the CRM immediately.

If you have questions about the amber API or need support setting up your workflow, contact our support team.