OpenClaw / Agent Plan reference

OpenClaw integration reference

Route OpenClaw's model provider and primary agent model through the Agent Plan Chat Completions endpoint.

1 · Quick start

Connect OpenClaw in three moves

Start with the matching configuration file, keep the key in the named secret location, and use one exact Model ID throughout the setup.

01

Open the OpenClaw configuration

Edit ~/.openclaw/openclaw.json and keep a backup of the existing provider and agent sections before merging changes.

02

Register the provider

Add the /v3 base URL, the delivered key, and the exact Model ID under models.providers. The API key should be treated as secret configuration.

03

Set the primary route

Point agents.defaults.model.primary to the same provider/model pair so OpenClaw does not register one model and call another.

The key is delivered to your Dashboard after payment. Replace the placeholder in the example, then keep the real value in your environment or local secret file.

2 · Configuration

Copy the provider blueprint

This is a safe starting point for the current Agent Plan route. Keep existing tool settings that are unrelated to the provider and replace only the placeholder key and Model ID values.

{
  "models": {
    "providers": {
      "volcengine": {
        "baseUrl": "https://ark.cn-beijing.volces.com/api/plan/v3",
        "apiKey": "YOUR_AGENT_PLAN_KEY",
        "models": [
          { "id": "glm-5.2", "name": "Agent Plan default model" }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": { "primary": "volcengine/glm-5.2" }
    }
  }
}

The default Model ID shown here is the stocked plan's current default. If you choose another supported model, replace every model field required by OpenClaw, not just the visible label.

Install / launch

Use the OpenClaw installation already documented for your environment.

3 · Verify the route

Confirm the connection before a full project

Restart the OpenClaw process after changing openclaw.json.

Run one small tool-use or chat task and confirm that the provider responds before adding automations.

Inspect the active provider and primary route if a model-not-found error appears.

Troubleshooting

Common configuration pitfalls

Merge these objects with your existing config. Replacing the whole file can remove unrelated agents or channels.

Use the Chat Completions-compatible /v3 route for this setup, not the Anthropic-compatible base URL.