Hermes Agent

Nous Research's open-source terminal AI agent: 60+ built-in tools, persistent memory, a Skills system, and messaging gateways for Telegram / Discord. It supports custom OpenAI-compatible providers, so it points straight at ByteSeek.

Install

Script install

Desktop

Linux / macOS / WSL2 / Android (Termux)

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash

Windows (native, PowerShell)

iex (irm https://hermes-agent.nousresearch.com/install.ps1)

The install script handles Python, Node.js and other dependencies; afterwards run source ~/.bashrc (or ~/.zshrc) to refresh commands.

Prerequisites

Hermes installed (hermes --version prints a version)
An OpenAI-format group key from the API Keys page

Configure

Run the setup wizard:

hermes setup

At the provider step choose the custom / OpenAI-compatible entry and fill in:

Base URL: https://tokens.byteseek.ai/v1
API Key: your sk- key
Model: an exact ID from the model plaza, e.g. gpt-5.6-sol

The wizard asks in order

Prompt	Value
Provider type	`custom` (custom OpenAI-compatible provider)
Base URL	`https://tokens.byteseek.ai/v1`
API Key	Your `sk-` key
Default model	A model ID, e.g. `gpt-5.6-sol`

The wizard then writes the config and returns to the main UI; a successful test message means you are connected.

Verify

hermes doctor
hermes chat -q "reply with the single word: ok"

doctor clean and chat replying normally means success. Charges appear on the usage page.

Common issues

Symptom	Fix
`hermes` not found after install	Shell not refreshed — `source ~/.bashrc` or open a new terminal
401	Wrong key — rerun `hermes setup` to update credentials
403	Model not in group, see Error Codes
Switching models	Run `hermes model`; prefix model IDs when using a composite key

#Hermes Agent

#Install

#Prerequisites

#Configure

#Verify