huawei-cloud-modelarts-skill
Manage Huawei Cloud ModelArts Resources
Simplify AI model development and deployment on Huawei Cloud ModelArts. This skill provides unified resource management for training jobs, models, inference services, and notebooks with built-in security features.
Download the skill ZIP
Upload in Claude
Go to Settings → Capabilities → Skills → Upload skill
Toggle on and start using
Test it
Using "huawei-cloud-modelarts-skill". List my training jobs
Expected outcome:
- Training Job: image-classification-v2
- Status: Running
- Created: 2026-04-21 10:30:00
- Training Progress: 75%
Using "huawei-cloud-modelarts-skill". Check resource overview
Expected outcome:
- Resource Overview:
- - OBS Buckets: 5
- - Active Training Jobs: 3
- - Deployed Models: 12
- - Running Services: 4
- - Active Notebooks: 2
Security Audit
SafeStatic analysis detected 35 potential issues, all are false positives after contextual review. Code execution patterns are markdown documentation examples, network URLs are legitimate documentation links, and credential handling implements data redaction (a security feature). The skill uses official modelarts-sdk with proper security practices: no credential storage, runtime-only authentication, and automatic sensitive data masking.
Risk Factors
⚙️ External commands (2)
🌐 Network access (3)
🔑 Env variables (1)
Quality Score
What You Can Build
Monitor Training Resources
Data scientist needs to check active training jobs and resource allocation across ModelArts platform
Deploy ML Models
DevOps engineer wants to deploy trained models as inference services and monitor service health
Manage Notebooks
Researcher needs to list and manage multiple ModelArts Notebook instances for different experiments
Try These Prompts
List all my ModelArts resources including training jobs, models, and services
Create a training job named my-model-training with code in obs://my-bucket/code/ and boot file train.py
Show me the last 10 models deployed on ModelArts
List all active inference services and their status
Best Practices
- Always verify credentials are properly configured before running resource operations
- Use descriptive job names when creating training jobs for easier tracking
- Check service status before deploying new models to avoid resource conflicts
- Monitor resource usage to optimize costs on ModelArts platform
Avoid
- Do not hardcode credentials in scripts or configuration files
- Avoid creating training jobs without proper resource limits
- Do not ignore error messages from API calls
- Never commit sensitive data or model artifacts to public repositories