How to use TrueConf chatbots with the Hermes AI agent
What is an AI agent and what are its benefits?
An AI agent is a program that either operates autonomously or assists a user upon request. The key advantage of this solution is that you can interact with it as if speaking to a human – there is no need to use commands with multiple parameters. This software understands almost any natural language, forwards the query to an AI model, and delivers the model’s response.
Hermes is one of the popular free AI agents. This tool supports a wide range of skills, and is able to launch subagents. It combines the power of language models and can perform various actions on the device where it is installed, for example, this solution can work with files, launch applications, or even write code. You can find more information on the Hermes official website.
Please keep in mind that TrueConf does not advocate the use of Hermes, does not recommend its deployment in a corporate infrastructure, and assumes no responsibility for the work of Hermes, related AI agents, external services, or third‑party components.
The plugin is used and connected to Hermes by a user or system administrator at their own discretion and with due consideration of possible risks. Before deploying this software in a corporate network, you must learn about security risks, including data leakage, unauthorised access, prompt injection, execution of unwanted commands, compromise of API keys, and unexpected behaviour of AI agents:
The plugin should be used only after you carry out a full risk assessment, test the software in an isolated environment, restrict access rights, configure monitoring, and take measures to protect the corporate infrastructure.
We recommend deploying the agent on a dedicated server or virtual machine to make sure that it is available 24/7. The main requirement is that this server has to be accessible from your local network. We will install Hermes on a dedicated VM, and users will interact with the agent in the chat of TrueConf client applications.
Getting ready
We will use Debian 13 to exemplify the integration of Hermes and TrueConf. Before starting installation, we recommend updating OS packages:
|
1 |
sudo apt update && sudo apt upgrade -y |
Install the curl tool if you do not have it:
|
1 |
sudo apt install curl -y |
Installation of Hermes
Choosing an AI Model
Before installation, you need to select an AI model and get an API key. To get an API key, select a provider. We will demonstrate the configuration using the free models provided by OpenCode. On the main page, choose Zen → Get started with Zen, and log in to your personal account. You will then see your API key, and below there will be a large list of available models (check only those marked Free):
Installation of the Hermes AI agent
To install Hermes, just run the following command:
|
1 |
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash |
The setup wizard will open. Below we explain each step in more detail.
The installation wizard will check whether your OS meets the requirements and install any missing components. Select Full setup in the menu How would you like to set up Hermes?
Now, let us discuss the main stages. First, select the AI model provider – in our case, OpenCode → OpenCode Zen:
Next paste the API key for accessing the model; leave the URL at its default value. We showed how to get the API key earlier:
We chose the model mimo-v2.5-free as an example:
When you see the prompt Select terminal backend, choose Local:
During the step Select platforms to configure, press Esc, because we will later configure our TrueConf plugin as the channel:
Also press Esc when you are in the Tools for CLI menu:
During the next step, select the recommended option Local Browser:
Select Skip when configuring image generation settings:
Choose Skip during the TTS settings step:
To let the AI agent search for information on the Internet, select a search provider, for example DuckDuckGo:
After configuration, run the command:
|
1 |
source ~/.bashrc |
Installation of the TrueConf plugin for Hermes
The adapter for the TrueConf messenger is installed directly from the corporate GitHub: the plugin project can be found here. To install the plugin, run this command:
|
1 |
curl -fsSL https://raw.githubusercontent.com/TrueConf/hermes-trueconf-adapter/main/install.sh -o /tmp/install.sh && bash /tmp/install.sh |
Let us go through each of the configuration steps in detail:
- Enter the server address
- Specify the login of the account under which the agent will work
- Enter the password for the agent account
- Choose whether you want to use SSL/HTTPS
- If you are using a self‑signed SSL certificate, disable SSL certificate verification (this may also be useful during testing)
- Allow users to write messages to your bot
- To let users send messages to the agent, do the following:
- Restart the gateway with the command:
- Send a message to the bot from one of TrueConf client applications, the bot will send you a pairing code in response.
- Enter the pairing code by running this command:
|
1 |
hermes gateway restart |
|
1 |
hermes pairing approve trueconf PAIR_CODE |
where PAIR_CODE should be replaced with the pairing code sent by the bot:
The user connected to TrueConf Server as the agent will have the Online status, and you can interact with it via the corporate chat in your TrueConf client application.
If necessary, you can edit the settings of the TrueConf adapter, for example, when you want to use a different server account for the AI agent. To edit settings, run this command:
|
1 |
hermes setup |
Skip the first steps until you reach the step Select platforms to configure, the TrueConf adapter will be listed there as TrueConf (configured):
Select the TrueConf adapter and change the required parameters by following the instructions in the adapter setup wizard.
You are now ready to start working with the AI agent.
Alexey Arkatov
Date published: 06.07.2026