Setup & Installation
openclaw skills install @carol-gutianle/modelready
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install modelready
What This Skill Does
ModelReady starts a local or Hugging Face model as an OpenAI-compatible server and lets you chat with it without leaving your conversation. It uses vLLM to serve the model on a port you specify.
Eliminates switching to a terminal to launch vLLM, track the process, and manually curl the endpoint just to send a message.
When to use it
- Testing a freshly downloaded Hugging Face model checkpoint
- Chatting with a local Qwen or LLaMA model mid-session
- Running a fine-tuned model on a specific port for a quick demo
- Checking whether a model server is still up during a long session
- Switching between two local models by starting them on different ports
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: start Qwen2.5-7B-Instruct on port 19001 and send it a greeting
AGENT
- 1Run /modelready start repo=Qwen/Qwen2.5-7B-Instruct port=19001
- 2Wait for the vLLM server to initialize and confirm it is ready
- 3Run /modelready chat port=19001 text="hello"
- 4Return the model response inline in the conversation
OUTPUT
The model's reply to "hello" displayed in chat
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.
vLLM installed on the host machineURL environment variable set to the server host address