ModelScope is an open-source platform for machine learning model sharing, evaluation, and deployment within automated agent workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install modelscope
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 modelscope 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 ModelScope?
ModelScope provides a collaborative environment for AI developers and researchers to discover, evaluate, and deploy a vast array of pre-trained models. This skill bridges the ModelScope platform with your development environment, enabling Openclaw Skills to interact with sophisticated machine learning pipelines and datasets seamlessly. By leveraging Membrane as the underlying transport layer, it removes the complexity of authentication and infrastructure management.
Using this integration, developers can streamline the entire AI development lifecycle. Whether you are retrieving a specific model for sentiment analysis or managing large-scale datasets for fine-tuning, this skill ensures that your Openclaw Skills have direct, secure access to the ModelScope ecosystem.
ModelScope Use Cases
- Discovering and deploying pre-trained machine learning models for specialized AI tasks.
- Managing and versioning datasets used for training and research collaboration.
- Automating model evaluation workflows using standardized pipelines.
- Integrating high-performance AI models into custom Openclaw Skills without manual credential handling.
How ModelScope Works
- Initialize the environment by installing the Membrane CLI to facilitate communication between the agent and the ModelScope API.
- Perform a secure login that associates your agent session with Openclaw Skills and handles tenant-based authentication.
- Create a connection to the ModelScope domain, which triggers a server-side authentication flow to manage tokens securely.
- Search for specific ModelScope actions using natural language intents to find the right model or dataset tools.
- Execute model pipelines or retrieve data objects by running specific actions through the CLI or proxying direct API requests.
ModelScope Setup
First, install the necessary CLI tool:
npm install -g @membranehq/cli@latest
Authenticate your agent, ensuring you specify your environment to optimize the integration for Openclaw Skills:
membrane login --tenant --clientName=openclaw
Establish a secure connection to the ModelScope platform:
membrane connection ensure "https://www.modelscope.cn/" --json
ModelScope Data Schema & Taxonomy
The ModelScope integration focuses on three primary data pillars to maintain organization and metadata clarity:
| Data Entity | Description |
|---|---|
| Model | Metadata and weights for pre-trained machine learning architectures. |
| Pipeline | Logical workflows used for model execution and inference tasks. |
| Dataset | Curated collections of data structured for training or evaluation. |
Each interaction through Openclaw Skills returns structured JSON objects including action IDs, input schemas for parameters, and output schemas for result parsing.
ModelScope Advanced Features
- Intent-based discovery that allows agents to find ModelScope actions using natural language descriptions.
- Automated credential refreshing via Membrane, ensuring Openclaw Skills maintain persistent access without local secret storage.
- Proxy request capability for direct ModelScope API access with automatic header and authentication injection.
- Support for headless environments, allowing for robust server-side automation and remote agent deployments.
SKILL.md
Loading
Related Openclaw Skills
The Things Network
Connect and manage LoRaWAN IoT devices and automate data workflows through The Things Network using AI agents.
Dataddo AI Integration
A powerful integration skill that enables AI agents to extract, transform, and load data via the Dataddo platform using automated authentication.
Spacelift
A powerful integration for managing Spacelift infrastructure-as-code deployments and workflows through an AI-ready CLI interface.
Statuspal
A comprehensive integration for managing Statuspal status pages, incident reporting, and service health monitoring through automated workflows.
OpenAI
A comprehensive integration for managing OpenAI Assistants, conversation threads, and file storage through a secure, managed authentication layer.
Jobber
A comprehensive integration for field service management that allows AI agents to automate scheduling, invoicing, and customer management within Jobber.
Featured*
