Skip to main content

​
What is OpenClaw?

OpenClaw is a personal AI assistant that runs on your own device. It connects to various messaging platforms (Telegram, Discord, Slack, WhatsApp, etc.) and can be configured to use GPTProto’s powerful AI models through our API.
With GPTProto’s unified API, OpenClaw can access 160+ AI models from top providers including OpenAI, Google, Anthropic, DeepSeek, and more — all with a single API key.

​
Install and Configure OpenClaw

1

Get Your GPTProto API Key

  1. Visit gptproto.com and sign up or log in.
  2. Navigate to the Dashboard and create a new API Key (sk-xxxxx).
  3. Keep this key secure — you’ll need it in the next step.
Keep your API key secure and never share it publicly. Treat it like a password.
2

Install OpenClaw

For detailed installation guides, refer to the official documentation.
Prerequisites: Node.js 22 or newerThe easiest way to install OpenClaw is using the official install script:macOS/Linux:
curl -fsSL https://openclaw.ai/install.sh | bash
Windows (PowerShell):
iwr -useb https://openclaw.ai/install.ps1 | iex
3

Set Up OpenClaw

After running the install command, the setup wizard will start automatically. If it doesn’t, run:
openclaw onboard --install-daemon
If you’ve already initialized before, you can run openclaw config and select model to reconfigure.Begin setup:
  • I understand this is powerful and inherently risky. Continue? | Select ● Yes
  • Onboarding mode | Select ● Quick Start
  • Model/auth provider | Select ● Skip for now
  • Filter models by provider | Select ● anthropic
  • Default model | Select ● Keep current
  • Complete the remaining configuration based on your own preferences.
Selecting Skip for now for the model/auth provider is important — it lets you manually configure GPTProto’s custom model parameters in the next step instead of being locked into a pre-configured provider.
4

Configure GPTProto Models

To configure a custom model, edit ~/.openclaw/openclaw.json. You can open it with any text editor:
nano ~/.openclaw/openclaw.json
1. Add GPTProto provider configurationAdd the following under models.providers. If you skipped onboarding, the models section may not exist yet — add it manually. Replace YOUR_GPTPROTO_KEY with the API key from Step 1.
{
  "models": {
    "mode": "merge",
    "providers": {
      "gptproto": {
        "baseUrl": "https://gptproto.com",
        "apiKey": "YOUR_GPTPROTO_KEY",
        "api": "anthropic-messages",
        "models": [
          {
            "id": "claude-opus-4-6",
            "name": "claude-opus-4-6",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          },
          {
            "id": "claude-sonnet-4-6",
            "name": "claude-sonnet-4-6",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          },
          {
            "id": "glm-5",
            "name": "glm-5",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          },
          {
            "id": "MiniMax-M2.5",
            "name": "MiniMax-M2.5",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          },
          {
            "id": "kimi-k2.5",
            "name": "kimi-k2.5",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          }
        ]
      }
    }
  }
}
2. Set the default agent model
{
  "agents": {
    "defaults": {
      "model": {
        "primary": "gptproto/claude-opus-4-6" // default model
      },
      "models": { // available models
        "gptproto/claude-opus-4-6": {},
        "gptproto/claude-sonnet-4-6": {},
        "gptproto/glm-5": {},
        "gptproto/MiniMax-M2.5": {},
        "gptproto/kimi-k2.5": {}
      }
    }
  }
}
Complete ~/.openclaw/openclaw.json example
{
  "meta": {...},
  "wizard": {...},
  "models": {
    "mode": "merge",
    "providers": {
      "gptproto": {
        "baseUrl": "https://gptproto.com",
        "apiKey": "YOUR_GPTPROTO_KEY",
        "api": "anthropic-messages",
        "models": [
          {
            "id": "claude-opus-4-6",
            "name": "claude-opus-4-6",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 200000,
            "maxTokens": 100000
          },
          ...
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "gptproto/claude-opus-4-6" // default model
      },
      "models": { // available models
        "gptproto/claude-opus-4-6": {},
        "gptproto/claude-sonnet-4-6": {},
        "gptproto/glm-5": {},
        "gptproto/MiniMax-M2.5": {},
        "gptproto/kimi-k2.5": {}
      },
      ...
    }
  },
  ...
}
5

Complete Setup

Continue with the remaining OpenClaw configuration:
  • Select channel | Choose and configure your preferred messaging channel.
  • Configure skills | Select and install the skills you need.
  • Complete the setup.
6

Chat with Your Bot

After setup, the CLI will ask How do you want to hatch your bot?
  • Select ● Hatch in TUI (recommended)
Now you can start chatting with your AI bot in the Terminal UI!OpenClaw supports many channels for interaction, including Web UI, Discord, Slack, Telegram, and more. Refer to the official docs for channel setup: Channels Setup
  • For Web UI, open the Web UI (with token) link shown in your terminal.
7

Verify Installation

Verify everything is working correctly:
openclaw doctor     # Check configuration issues
openclaw status     # View gateway status
openclaw dashboard  # Open Dashboard in browser
For detailed configuration guides, refer to the official documentation.
OpenClaw may involve security risks if improperly configured or deployed without proper access controls. Refer to the official security documentation.

​
Advanced Configuration

​
Using Skills

Skills are folders containing a SKILL.md file. If you want to add new capabilities to your OpenClaw agent, ClawHub is the best way to find and install skills.

​
Install ClawHub

npm i -g clawhub

​
Manage Skills

Search for skills: 
clawhub search "postgres backups"
Install a new skill: 
clawhub install my-skill-pack
Update installed skills: 
clawhub update --all

​
Plugin Management

Plugins are small code modules that extend OpenClaw with additional capabilities (commands, tools, and Gateway RPC).
View loaded plugins: 
openclaw plugins list
Install official plugins (example: voice-call): 
openclaw plugins install @openclaw/voice-call
Restart the gateway: 
openclaw gateway restart

​
Why GPTProto + OpenClaw?

FeatureBenefit
160+ ModelsAccess OpenAI, Google, Anthropic, DeepSeek, and more with one API key
Unified APIOpenAI-compatible format — no code changes when switching models
Auto FailoverSeamlessly switch between providers when one is unavailable
Cost EfficientPay-as-you-go pricing, compare models and choose the best value
Multi-ChannelConnect to Telegram, Discord, Slack, WhatsApp, and more
ExtensibleAdd skills and plugins from ClawHub for endless capabilities

​
Troubleshooting

​
Common Issues

  • API Key Authentication
    • Ensure your GPTProto API key is valid and has sufficient balance
    • Check that the API key is correctly set in your configuration
    • Verify at gptproto.com/dashboard
  • Model Availability
  • Connection Issues
    • Ensure the OpenClaw gateway is running: openclaw status
    • Check network connectivity to gptproto.com
    • Run openclaw doctor for diagnostics

​
Resources