OpenClaw Integration

This guide shows how to configure OpenClaw to use model capabilities through the unified R9S API.

Prerequisites

Before you start, make sure you have:

OpenClaw installed

OpenClaw recommends Node 24, or Node 22.14 and later. The installer script can handle Node setup automatically.

macOS, Linux, or WSL2:
Terminal window
```
curl -fsSL https://openclaw.ai/install.sh | bash
```
Windows PowerShell:
Terminal window
```
iwr -useb https://openclaw.ai/install.ps1 | iex
```
An R9S API key

Sign in to the R9S platform, create an API key, and copy it.

Start the onboarding wizard:

openclaw onboard

During onboarding, configure OpenClaw as follows.

Choose Custom Provider or Other (OpenAI Compatible API).

Use the following R9S API settings:

Field	Value
Base URL	`https://api.r9s.ai/v1`
API key input	Choose Enter API key directly, then paste your R9S API key

After successful verification, OpenClaw generates an endpoint ID, such as custom-api-r9s-ai, and continues to the next step.

Optionally enter a memorable alias, such as r9s, to make the provider easier to select later.

Press Enter to skip this step if you do not need an alias.

Enter the model ID you want to use through R9S, for example:

gpt-4.1

or:

claude-3-haiku

When you see Verification successful, OpenClaw has connected to R9S successfully, and the Base URL and API key are valid.

When OpenClaw shows the skill dependency list, select Skip for now unless you specifically need those skills.

OpenClaw may ask for additional keys such as GOOGLE_PLACES_API_KEY for specific skills.

These keys are optional and are not required for R9S model chat. You can choose No or press Enter to skip them.

Select Hatch in TUI to start OpenClaw in the terminal UI. This is the recommended command-line mode.

In the OpenClaw TUI chat interface, check the bottom status bar. You should see a provider and model similar to:

custom-api-r9s-ai/gpt-4.1

This means OpenClaw is using gpt-4.1 through R9S.

Send a test message, such as:

hello

If OpenClaw returns a model response, the integration is working.

OpenClaw reached R9S, but R9S rejected the request. This usually means one of the following values needs to be checked:

The request path is incorrect. Check that the Base URL is exactly:

https://api.r9s.ai/v1

If OpenClaw appends its own path to the provider URL, a missing or extra path segment can cause a 404 error.