LLM task

is an optional plugin tool that runs a JSON-only LLM task and returns structured output (optionally validated against JSON Schema).

This is ideal for workflow engines like Lobster: you can add a single LLM step without writing custom OpenClaw code for each workflow.

Enable the plugin

1. Enable the plugin:

json
Copy
{
  "plugins": {
    "entries": {
      "llm-task": { "enabled": true }
    }
  }
}

2. Allowlist the tool (it is registered with
   text

   Copy

   optional: true
   ):

json
Copy
{
  "agents": {
    "list": [
      {
        "id": "main",
        "tools": { "allow": ["llm-task"] }
      }
    ]
  }
}

Config (optional)

json
Copy
{
  "plugins": {
    "entries": {
      "llm-task": {
        "enabled": true,
        "config": {
          "defaultProvider": "openai-codex",
          "defaultModel": "gpt-5.5",
          "defaultAuthProfileId": "main",
          "allowedModels": ["openai/gpt-5.4"],
          "maxTokens": 800,
          "timeoutMs": 30000
        }
      }
    }
  }
}

is an allowlist of strings. If set, any request outside the list is rejected.

Tool parameters

• text

  Copy

  prompt
  (string, required)
• text

  Copy

  input
  (any, optional)
• text

  Copy

  schema
  (object, optional JSON Schema)
• text

  Copy

  provider
  (string, optional)
• text

  Copy

  model
  (string, optional)
• text

  Copy

  thinking
  (string, optional)
• text

  Copy

  authProfileId
  (string, optional)
• text

  Copy

  temperature
  (number, optional)
• text

  Copy

  maxTokens
  (number, optional)
• text

  Copy

  timeoutMs
  (number, optional)

accepts the standard OpenClaw reasoning presets, such as or .

Output

Returns

containing the parsed JSON (and validates against when provided).

Example: Lobster workflow step

lobster
Copy
openclaw.invoke --tool llm-task --action json --args-json '{
  "prompt": "Given the input email, return intent and draft.",
  "thinking": "low",
  "input": {
    "subject": "Hello",
    "body": "Can you help?"
  },
  "schema": {
    "type": "object",
    "properties": {
      "intent": { "type": "string" },
      "draft": { "type": "string" }
    },
    "required": ["intent", "draft"],
    "additionalProperties": false
  }
}'

Safety notes

• The tool is JSON-only and instructs the model to output only JSON (no code fences, no commentary).
• No tools are exposed to the model for this run.
• Treat output as untrusted unless you validate with
  text

  Copy

  schema
  .
• Put approvals before any side-effecting step (send, post, exec).

Related

• Thinking levels
• Sub-agents
• Slash commands