OpenClaw with TokenPass
Install OpenClaw through the official project and configure TokenPass as the model provider.
1. Install OpenClaw
macOS, Linux, or WSL:
bash
curl -fsSL --proto '=https' --tlsv1.2 https://openclaw.ai/install.sh | bashWindows PowerShell:
powershell
iwr -useb https://openclaw.ai/install.ps1 | iexAlternatively, install with npm:
bash
npm install -g openclaw@latest
openclaw onboard --install-daemon2. Configure TokenPass
During onboarding, choose a custom OpenAI-compatible or Anthropic-compatible provider according to the selected model. Enter:
| Setting | Value |
|---|---|
| API key | Your TokenPass API key |
| Base URL | https://tokenpass.cloud/v1 |
| Model | An ID returned by GET /v1/models |
3. Verify the connection
bash
openclaw gateway status
openclaw doctorStart a session and send a short message. If the model is unavailable, refresh the model ID from TokenPass before changing the OpenClaw installation.
Security
OpenClaw can execute tools and connect to messaging channels. Keep its gateway private, use pairing and allowlists, and review the official security guide before exposing it remotely.