User GuideAPI ReferenceAI Applications

OpenClaw

OpenClaw is an open-source, self-hosted personal AI agent and gateway platform. It seamlessly connects multi-channel messages from Telegram, Discord, WhatsApp, and more to AI agents running on local hardware, supporting multi-agent cooperative routing, persistent memory, and autonomous tool calling, allowing developers to take full control of their private AI assistants.

✨ AI One-Click Configuration

Copy the prompt below and send it to your AI assistant to automatically detect your environment, install, and configure OpenClaw:

AI 配置提示词 (Prompt)
点击卡片任意位置复制
Please refer to the OpenRely official documentation (https://openrely.ai/en/docs/apps/openclaw) to help me configure OpenClaw on this machine:
1. Obtain API Key:If I have not provided an API key yet, please remind me to visit the OpenRely Console (https://api.openrely.ai/keys) to create and enter it.
2. Environment Installation:Check if OpenClaw is installed locally. If not, run curl -fsSL https://openclaw.ai/install.sh | bash and execute openclaw onboard.
3. Configuration File:Modify ~/.openclaw/openclaw.json to add the openrely provider under models.providers:
Set baseUrl to https://api.openrely.ai/v1, api to openai-completions, and apiKey to <MY_API_KEY>;
Set agents.defaults.model.primary to the primary model (e.g., openrely/<primary_model_id>; you can check the supported models at any time on the OpenRely Model Pricing page https://api.openrely.ai/pricing).
4. Verify Readiness:Run openclaw gateway status to check the service status, and execute openclaw dashboard to open the console for verification.

🔑 Prerequisites


🚀 Steps

Install and Initialize OpenClaw

OpenClaw supports one-click installation on macOS and Linux:

# 1. Run the official one-click installation script
curl -fsSL https://openclaw.ai/install.sh | bash

# 2. Run the basic onboarding wizard
openclaw onboard --install-daemon

The wizard will complete basic gateway process initialization and daemon registration.

Configure OpenRely Endpoint and Models

OpenClaw uses ~/.openclaw/openclaw.json to manage multi-model gateway configurations. Open this file and declare the openrely custom provider under models.providers:

{
  models: {
    mode: "merge",
    providers: {
      openrely: {
        baseUrl: "https://api.openrely.ai/v1",
        apiKey: "sk-your-openrely-api-key", // Or use environment variable "${OPENRELY_API_KEY}"
        api: "openai-completions",
        models: [
          { id: "<primary_model_id>", name: "Primary Model" },
          { id: "<fallback_model_id>", name: "Fallback Model" }
        ],
      },
    },
  },
  agents: {
    defaults: {
      model: {
        primary: "openrely/<primary_model_id>",
        fallbacks: ["openrely/<fallback_model_id>"],
      },
    },
  },
}

Start and Verify

Check the OpenClaw gateway daemon status and open the web management dashboard:

# 1. Check gateway service status
openclaw gateway status

# 2. Open visual management dashboard
openclaw dashboard

You can also run the following command in your terminal to verify that the loaded model list includes models prefixed with openrely/:

openclaw models list

If you can successfully initiate a conversation and receive streaming responses in the dashboard, the integration with OpenRely is complete.


💡 Common Commands

Frequently used control commands for OpenClaw operations and management:

CommandDescription
openclaw dashboardLaunch the browser and open the visual web management dashboard
openclaw gateway statusView the current gateway daemon status and listening ports
openclaw gateway restartRestart the gateway process to quickly reload the latest configuration file
openclaw models listList all registered models and provider statuses

❓ FAQ

Encountering other API errors or billing questions? Please refer to the complete FAQ and Troubleshooting Guide.

Was this page helpful?

OpenClaw | OpenRely Docs