worker-model-switch

Switch a Worker Agent's LLM model by updating MinIO configuration and notifying reload.

5.3k|652|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/alibaba/hiclaw --skill worker-model-switch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: worker-model-switch
Source: https://github.com/alibaba/hiclaw/tree/main/manager/agent/skills/worker-model-switch
Command: npx skills add https://github.com/alibaba/hiclaw --skill worker-model-switch

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill allows for the seamless modification of a Worker Agent's Large Language Model (LLM) configuration, enabling dynamic adaptation to different task requirements or performance needs.

Core Features & Use Cases

  • Dynamic Model Swapping: Change the LLM used by a Worker Agent without manual intervention.
  • Configuration Management: Updates the Worker's configuration in persistent storage (MinIO) and notifies the agent to reload.
  • Use Case: When a new, more capable LLM is released, or if a specific task requires a model with a larger context window, an administrator can use this Skill to switch the relevant Worker Agent to the new model.

Quick Start

Use the worker-model-switch skill to change the model for the worker named 'alice' to 'claude-sonnet-4-6'.

Frequently Asked Questions about worker-model-switch

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I switch the LLM model for a Worker Agent dynamically?▼

To switch an LLM model for a Worker Agent, this skill updates the worker's configuration in MinIO, patches openclaw.json, updates workers-registry.json, and sends a Matrix notification to trigger a file-sync and agent reload.

How does Worker Agent model switching handle configuration updates?▼

Worker Agent model switching resolves model parameters, tests AI gateway connectivity, patches the worker's openclaw.json, and updates workers-registry.json in MinIO to ensure persistent storage reflects the new LLM configuration.

Do I need MinIO to change a Worker Agent's LLM configuration?▼

Yes, MinIO is required because the skill updates the worker's configuration by patching openclaw.json and workers-registry.json directly within MinIO persistent storage before notifying the agent.

Can I use this skill to adapt a Worker Agent for a larger context window?▼

Yes, you can use this skill to dynamically adapt a Worker Agent to a new model with a larger context window by resolving the new model parameters and patching the worker's configuration.

Why does model switching send a Matrix notification?▼

Switching a model sends a Matrix notification to trigger file-sync, ensuring the Worker Agent receives the updated openclaw.json configuration from MinIO and reloads the new LLM seamlessly.

What is the best way to update a Worker Agent to a newly released LLM?▼

The best way to update a Worker Agent to a newly released LLM is using this skill to resolve model parameters, test AI gateway connectivity, patch MinIO configurations, and notify the agent to reload.