Introduction
DevOps engineers and developers know that managing Hetzner cloud infrastructure involves dozens of routine operations per day. Creating instances, configuring networks, deploying applications, monitoring load — each task requires either manual console access or writing bash scripts. According to a 2025 DevOps community survey, up to 40% of engineers' time is spent on such operations (source: State of DevOps Report 2025, Puppet).
The ASI Biont AI agent changes this paradigm. Now you can connect Hetzner via API, give the agent your API key in the chat, and it will write the integration code, execute commands, and set up monitoring on its own. No control panels, no "add integration" buttons — everything is done through a dialogue with the AI. This reduces time spent on routine tasks by up to 70% and allows you to focus on architecture and optimization.
What is Hetzner and Why Connect It to an AI Agent
Hetzner is one of the largest European cloud providers, offering virtual servers (VPS), dedicated servers, and object storage. Its API (documentation: docs.hetzner.cloud) allows you to manage all resources programmatically: create servers, configure networks, monitor metrics, manage DNS.
Connecting ASI Biont to Hetzner via the API enables you to:
- Give voice or text commands in natural language (e.g., "spin up a server with Ubuntu 24.04, 4 vCPU, 8 GB RAM")
- Automate application deployment without manually writing scripts
- Monitor performance and receive failure notifications
- Manage instances without direct SSH access or the Hetzner console
How ASI Biont Connects to Hetzner
The connection process is extremely simple and requires no technical preparation:
- Log into your Hetzner Cloud account and generate an API token (under API Tokens in the control panel). A token with read and write permissions for all projects is sufficient.
- Open the chat with the ASI Biont AI agent at asibiont.com.
- Write: "Connect my Hetzner, here is the API key: [your token]".
ASI Biont will independently read the Hetzner API documentation, write the integration code in Python (using the requests library), establish a secure connection, and verify access. The entire process takes a few seconds. After that, you can give commands in natural language.
Important: The AI agent does not store your API key in plain text — it is encrypted and used only to execute your commands. You can revoke the token at any time in the Hetzner panel.
What Tasks Does the Integration Automate
The integration covers all major operations with Hetzner cloud infrastructure. Here is a table of key capabilities:
| Task Type | Example Command | Result |
|---|---|---|
| Server creation | "Create a server in Helsinki with 2 vCPU, 4 GB RAM, OS Ubuntu 24.04" | ASI sends a POST request to /servers, gets the IP and password, and displays them in the chat |
| Application deployment | "Deploy my Node.js project from the repository github.com/user/app to server 192.168.1.10" | The AI generates SSH commands, clones the repository, installs dependencies (npm install), and starts the application via PM2 |
| Monitoring | "Show CPU and RAM usage for all servers" | ASI sends requests to /servers and /actions, parses the metrics (available via the monitoring API), and displays a table |
| Scaling | "If CPU > 80% on server web-01, create another server and add it to the load balancer" | The AI creates a rule: periodically poll metrics, and when exceeded, send a request to create a server and update the load balancer (requires configuration via the Hetzner Load Balancer API) |
| DNS management | "Add an A record api.example.com -> 1.2.3.4" | ASI sends a request to the Hetzner DNS API (via their DNS service) and creates the record |
| Backups | "Set up automatic daily backups for all servers" | The AI enables the backup option in the server configuration (via the backups: true parameter) |
Examples of Specific Use Cases
Scenario 1: Quick Deployment of a Test Environment
A developer urgently needs to test a new version of an application. Instead of going to the Hetzner console, creating a server, configuring SSH, and cloning the code, they simply write in the ASI Biont chat:
"Create a server in Nuremberg with 2 vCPU, 4 GB RAM, Ubuntu 24.04, open ports 80 and 443, install the latest version of Nginx, deploy my site from the GitHub repository myorg/myapp, and bind the domain test.myapp.com via DNS"
The AI agent executes the following sequence:
1. Creates a server via POST /servers (selects the image, type, location)
2. Waits for the server to reach the running status (polls via GET /servers/{id})
3. Opens ports in the firewall (via the Firewalls API)
4. Connects via SSH (using the generated password) and installs Nginx
5. Clones the repository and configures the web server
6. Creates a DNS record via the Hetzner DNS API
7. Returns the link: "The site is available at http://test.myapp.com"
All this takes 2-3 minutes instead of 15-20 when done manually.
Scenario 2: Monitoring and Automatic Scaling
A system administrator needs to monitor the load on production servers. They set up a rule:
"Monitor CPU usage on server prod-01. If the average load over 5 minutes exceeds 75%, create a new server with the same specifications and add it to the load balancer load-balancer-01. When the load drops below 30%, remove the extra server"
ASI Biont creates a monitoring script that:
- Polls CPU metrics via the Hetzner monitoring API every 60 seconds
- When the threshold is exceeded, sends a request to create a server (with the same image, type, and SSH keys)
- Adds the server to the load balancer via the Load Balancer API
- When the load decreases, removes the server (after verifying that traffic has been redistributed)
This allows automatic adaptation to peak loads without human intervention.
Scenario 3: Audits and Reports
A team needs to report weekly on the state of the infrastructure. Instead of manually collecting data, they can simply request:
"Generate a report for all servers: name, IP, status, CPU and RAM usage, cost for the current month. Send it to Slack"
The AI agent collects data from the Hetzner API (list of servers, metrics, pricing information from the price list), formats it into a table, and sends it via the Slack integration.
Technical Details: How the AI Writes Integration Code
When you provide the API key, ASI Biont does not just save it — it analyzes the structure of the Hetzner API based on the official documentation (the OpenAPI specification is available at: https://docs.hetzner.cloud). The AI agent uses this specification to generate Python code:
import requests
BASE_URL = "https://api.hetzner.cloud/v1"
HEADERS = {"Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json"}
def create_server(name, server_type, image, location):
payload = {
"name": name,
"server_type": server_type,
"image": image,
"location": location
}
response = requests.post(f"{BASE_URL}/servers", headers=HEADERS, json=payload)
return response.json()
The code is generated dynamically for each request. If you ask to do something not available in the API (e.g., "format the disk"), the AI will inform you of the limitation and suggest an alternative.
Why It's Beneficial: Time and Resource Savings
- Reduction in time spent on routine operations by up to 70%. Instead of 10-15 minutes to create and configure a server, it takes 2-3 minutes via chat.
- Fewer errors. The AI agent strictly follows the API documentation, eliminating human typos in configuration.
- Accessible to non-DevOps specialists. Managers or developers without console access can issue commands via chat without knowing SSH or bash.
- Flexibility. You are not limited to pre-built integrations — ASI Biont connects to any service with an API. If you need to manage Hetzner via Telegram or voice, just set it up in the dialogue.
How to Get Started
- Go to asibiont.com and create an account (free trial period).
- Generate an API token in the Hetzner Cloud panel (under Security -> API Tokens).
- In the chat with the AI agent, write: "Connect Hetzner, here is the key: [token]".
- Start managing your infrastructure by voice or text.
Conclusion
The integration of ASI Biont with Hetzner is not just about convenience. It represents a shift in the approach to infrastructure management: from manual labor to a dialogue with AI. You stop spending time writing scripts and logging into consoles — you simply say what needs to be done, and the AI executes it using the official API.
Try it yourself: connect your Hetzner to ASI Biont at asibiont.com and see how automation transforms your routine.
Comments