Skip to main content
Use AIsa in Hermes Agent with one model endpoint and one API key. AIsa endpoint: 
https://api.aisa.one/v1

​
Prerequisites

Before you start, make sure you have:
  • An AIsa API key
  • A terminal on macOS, Linux, or Windows WSL2

​
1. Install Hermes Agent

Run the official installer: 
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
Reload your shell: 
source ~/.zshrc
Verify the installation: 
hermes doctor

​
2. Add AIsa as the model provider

Run: 
hermes model
Hermes will ask you to choose a model provider. Select: 
Custom Model / OpenAI-compatible endpoint
Select Custom Model / OpenAI-compatible endpoint This lets Hermes connect to AIsa through AIsa’s OpenAI-compatible API.

​
2.1 Configure AIsa endpoint

When Hermes asks for the API endpoint and API key, enter: 
API base URL: https://api.aisa.one/v1
API key: YOUR_AISA_API_KEY
If the endpoint is valid, Hermes will show that it can access the model list from AIsa: 
Verified endpoint via https://api.aisa.one/v1/models
167 model(s) visible

​
2.2 Choose API mode

Hermes API compatibility mode prompt Pick mode 1. Auto-detect or 2. Chat Completions — both work, since the base URL ends in /v1. Mode 2 is recommended because it’s explicit and easier to debug.
ModeRequired base URLUse for
1. Auto-detecthttps://api.aisa.one/v1Resolves to Chat Completions for /v1 URLs
2. Chat Completionshttps://api.aisa.one/v1Standard OpenAI-compatible endpoints
3. Responses / Codexhttps://api.aisa.one/v1Codex CLI and reasoning models needing server-side state
4. Anthropic Messageshttps://api.aisa.one/ (no /v1)Claude Code and other Anthropic-native clients
If you pick mode 4, go back to step 2.1 and remove /v1 from the base URL — Anthropic clients append /v1/messages themselves.

​
2.3 Choose a model

Hermes will show the available model list. You can either enter the model number or type the model name directly: 
Select model [1-167] or type name: gpt-5
Choose a text, reasoning, or coding model. Do not choose image, video, embedding, Whisper, or TTS models as the main Hermes model.

​
2.4 Set context length

When Hermes asks: 
Context length in tokens [leave blank for auto-detect]:
Press Enter to leave it blank.

​
3. Connect AIsa capabilities and start using

After the model provider is configured, start Hermes: 
hermes
Then paste this prompt into Hermes: 
Run curl -sL https://aisa.one/docs/llms.txt and follow the returned instructions to connect AIsa capabilities.
Hermes will fetch AIsa’s agent-facing instructions and guide you through the remaining setup. After that, you can start using Hermes with AIsa models and capabilities. Try: 
Use AIsa capabilities to search the web and summarize the latest information about Hermes Agent.

​
Quick reference

ItemValue
Provider typeCustom Model / OpenAI-compatible endpoint
API base URLhttps://api.aisa.one/v1
API mode2. Chat Completions
Context lengthLeave blank for auto-detect
Capability setup promptRun curl -sL https://aisa.one/docs/llms.txt and follow the returned instructions to connect AIsa capabilities.

​
Troubleshooting

ProblemFix
hermes: command not foundRun source ~/.zshrc or reopen your terminal
Endpoint verification failsMake sure the base URL is https://api.aisa.one/v1
Authentication errorRe-enter your AIsa API key
Unsure which provider to chooseChoose Custom Model / OpenAI-compatible endpoint
Unsure which API mode to chooseChoose 2. Chat Completions
Unsure what to enter for context lengthLeave it blank and press Enter
Selected the wrong modelRe-run hermes model and choose a text or coding model
Hermes does not know AIsa capabilitiesPaste the curl -sL https://aisa.one/docs/llms.txt prompt into Hermes

Use AIsa in OpenClaw

Configure another agent runtime with the OpenClaw setup path.

Authentication

API key creation, rotation, and storage best practices.

Model Catalog

Browse supported models before choosing a Hermes default.