A powerful skill for executing local Large Language Model inference through the Ollama API using Python.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ollama-2
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install ollama-2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Ollama Local LLM Integration?
The Ollama skill for Openclaw Skills provides a seamless bridge between your local environment and powerful Large Language Models. By leveraging the Ollama API, this skill allows developers to trigger text generation, reasoning tasks, and model interactions directly from their terminal or automated workflows. It is designed for those who prioritize privacy, low latency, and the flexibility of running open-source models like Qwen locally.
This integration simplifies the complexity of managing local LLM endpoints. Whether you are building a private AI assistant or automating content generation, this Openclaw Skills module ensures your agent can communicate with specific model instances (such as qwen3.5:9b) without relying on expensive or data-leaking external cloud providers.
Ollama Local LLM Integration Use Cases
- Performing LLM inference tasks on local hardware to ensure data privacy.
- Automating text generation and creative writing via Python scripts.
- Interacting with specific model versions like qwen2.5 or qwen3.5 for consistent outputs.
- Developing and testing AI agent prompts without incurring API costs from cloud providers.
How Ollama Local LLM Integration Works
- The user initiates a request via the
ollama_query.pyscript with a specific prompt and optional model parameter. - The skill connects to the local or remote Ollama server (defaulting to http://100.66.1.2:11434).
- The request is processed by the Ollama instance using the specified model (e.g., qwen3.5:9b).
- The generated text response is returned to the Python script and outputted to the user or calling process.
Ollama Local LLM Integration Setup
To get started with this skill, ensure you have Ollama installed and running. Then, use the following commands:
# Install the skill via npx
npx skills add ayflying/ai-skills --skill ollama
# Install required Python dependencies
pip install requests
# Ensure your model is available locally
ollama pull qwen3.5:9b
# Run the Ollama service
ollama serve
Configure your environment variables by copying the provided .env.example to a new .env file.
Ollama Local LLM Integration Data Schema & Taxonomy
The Ollama skill maintains a structured directory to manage scripts and configurations:
| File/Folder | Description |
|---|---|
scripts/ollama_query.py |
Core logic for querying the Ollama API. |
scripts/test_ollama.py |
Connectivity utility to verify API and model status. |
.env.example |
Template for defining host address and default model names. |
SKILL.md |
Technical documentation and usage guidelines. |
assets/ |
Directory for storing related resources and images. |
Ollama Local LLM Integration Advanced Features
- Custom Host Support: Connect to remote Ollama instances across a local network or VPN.
- Model Flexibility: Dynamically switch between different models (e.g., Qwen, Llama, Mistral) at runtime via CLI arguments.
- Connection Testing: Includes a dedicated testing script to validate environment readiness before production use.
- Environment Variable Integration: Easily switch between development and production Ollama endpoints using .env configurations.
SKILL.md
Loading
Related Openclaw Skills
Carrier Logistics Index Analyzer
A sophisticated data analysis tool that extracts carrier performance metrics from spreadsheets to provide industry benchmarking and trend visualization.
Account API Skill
A comprehensive AI agent skill for managing financial account lifecycles, holder verifications, and legal documentation via the Adyen Account API.
Web3 Weekly Capital Report Generator
An automated data-fetching and reporting tool designed to generate comprehensive weekly insights into Web3 capital movements, regulatory shifts, and institutional activities.
Google Research Pro
An advanced web-searching skill that uses Playwright to bypass bot detection and deliver technical research directly to Telegram.
AI Task Management
A zero-dependency, high-performance task management skill designed for AI agents to handle the full lifecycle of project tasks with human oversight.
Mood CLI
A command-line sentiment analysis tool that converts text into weather-themed emotional insights and healing messages.
Featured*
