coala-client
How to use the coala-client CLI for chat with LLMs, MCP servers, and skills. Use when the user asks how to use coala, run coala chat, add MCP servers, import CWL toolsets, list or call MCP tools, or import or load skills.
Why use this skill?
Learn to use the coala-client CLI to chat with LLMs, manage MCP servers, import CWL toolsets, and load custom skills for your AI agent.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/hubentu/coalaWhat This Skill Does
The coala-client skill provides a powerful command-line interface for the coala ecosystem, enabling users to interact seamlessly with a variety of LLMs including OpenAI, Gemini, and Ollama. More than just a chat interface, it serves as an orchestration hub for the Model Context Protocol (MCP). This allows you to integrate local CWL (Common Workflow Language) toolsets, connect to remote MCP servers, and import specialized skills to extend the agent's capabilities. By managing configuration files in your home directory, coala-client acts as a bridge between your local data and large language models, providing a unified environment for complex technical workflows.
Installation
To integrate this functionality into your OpenClaw environment, run the following command in your terminal:
clawhub install openclaw/skills/skills/hubentu/coala
After installation, run coala init to generate the necessary directory structure at ~/.config/coala/ including your mcp_servers.json and env files. Ensure you have your LLM provider API keys exported in your shell environment or defined within the ~/.config/coala/env file for seamless connectivity.
Use Cases
This skill is designed for power users and developers who need to combine LLM reasoning with specific local tools or external data sources. You can use it to automate technical tasks by importing custom CWL workflows as MCP tools, making them available for the LLM to call during a chat session. It is also ideal for researchers needing to interface with specialized scientific databases via MCP or developers who want to manage multiple LLM backends through a single, consistent interface. You can even load modular skill sets dynamically into your chat context to handle specific domains of knowledge.
Example Prompts
- "How do I import a new CWL toolset from my local directory into the coala client?"
- "I want to use the OpenAI provider with the coala chat, how do I set that up and what is the command to list my active MCP servers?"
- "Load the data-analysis skill and then use the gene-variant.ncbi_datasets_gene tool with the following parameters: {"data": [{"gene": "TP53", "taxon": "human"}]}"
Tips & Limitations
To maintain security, be mindful that coala-client can execute tools and read local files based on your configurations. Regularly review your mcp_servers.json to ensure only trusted servers have access to your system. If you experience performance issues, use the --no-mcp flag to initiate a lighter chat session without tool overhead. Always check coala config to verify that your environment variables are correctly mapped. Note that while Ollama runs locally without keys, ensure your local instance is reachable at the correct base URL if you encounter connection errors.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-hubentu-coala": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution, external-api