agentsclimarketplace

Huawei cloud modelarts skill

Skill modelarts-agent/huawei-cloud-modelarts-skill

Huawei Cloud ModelArts platform integration. Modular design, OBS as primary storage support, environment variable authentication, includes pool plugin management and node configuration template queries.From its SKILL.md

Install
npx -y skills add modelarts-agent/huawei-cloud-modelarts-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

9.8 KB, ~2.2k tokens by cl100k_base, as published. Nobody here has run it

ModelArts Skill

Huawei Cloud ModelArts platform integration skill, featuring modular design, OBS as Notebook primary storage, and environment variable auto-authentication.


๐Ÿ” Security Model (CRITICAL)

All credentials handled internally, never returned to LLM:

โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
โ”‚                     LLM / OpenClaw                          โ”‚
โ”‚  โŒ  NEVER sees: AK, SK, Security Token, raw credentials     โ”‚
โ”‚  โœ…  ONLY sees: Masked status, API results, resource IDs     โ”‚
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
                            โ†• (Safe data only)
โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
โ”‚              auth_manager.py (Secure Vault)                 โ”‚
โ”‚  ๐Ÿ”’  Credentials stored in memory only                       โ”‚
โ”‚  โœ…  Only masked data exposed externally                      โ”‚
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
                            โ†• (Secure session)
โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
โ”‚              Module Functions (Secure SDK Wrapper)          โ”‚
โ”‚  Unified pattern: access.sdk().execute(api_func, ...)        โ”‚
โ”‚  Returns: Standard format API responses only                 โ”‚
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜

๐Ÿ”‘ Authentication Methods

ModeHow It Works
Notebook ModeAuto-detects ModelArts environment, zero configuration
Local ModeReads MODELARTS_AK/SK/PROJECT_ID/REGION environment variables

๐Ÿ“š Module Routing

Use keywords to find the corresponding reference file for function index and locations.

| Keyword | Reference File | |---------|----------------|---------------|--------| | notebook, jupyter, image, obs, flavor, cluster, auth | references/notebook.md | | pool, resource pool, dedicated pool, node, workload, os network, plugin | references/pool.md | | node pool, nodepool, scale node, node group, plugin, config template | references/node_pool.md | | train, training job, ่ฎญ็ปƒไปปๅŠก | references/train.md | | infer1.0, inference, service, ๆ—ง็‰ˆๆŽจ็†ๆœๅŠก,ๆ—ง็‰ˆๅœจ็บฟๆœๅŠก | references/infer_v1.md | | infer2.0, inference, service, ๆ–ฐ็‰ˆๆŽจ็†ๆœๅŠก๏ผŒๅœจ็บฟๆœๅŠก,ๆ–ฐ็‰ˆๅœจ็บฟๆœๅŠก | references/infer_v2.md | | management, workspace, authmode, authorization, quota, tag, scheduled event | references/management.md | | swr, image repo, ้•œๅƒไป“ๅบ“ | references/swr.md | | vpc, virtual private, subnet, security group, ่™šๆ‹Ÿ็งๆœ‰ไบ‘ | references/vpc.md | | kms, key management, ๅฏ†้’ฅ็ฎก็† | references/kms.md | | liteserver, dev server, hyper cluster, hyperinstance, eip, roce | references/liteserver.md | | common, auth, api helper, format result, decorator | references/common.md |


๐Ÿ“ฆ Module Architecture

modelarts/
โ”œโ”€โ”€ scripts/
โ”‚   โ”œโ”€โ”€ common_module/          # ่ทจๆจกๅ—ๅ…ฑไบซๅŸบ็ก€่ฎพๆ–ฝ
โ”‚   โ”‚   โ”œโ”€โ”€ __init__.py
โ”‚   โ”‚   โ”œโ”€โ”€ _bootstrap.py       # ็ปŸไธ€ๅผ•ๅฏผ
โ”‚   โ”‚   โ”œโ”€โ”€ auth.py             # ่ฎค่ฏ็ฎก็†
โ”‚   โ”‚   โ”œโ”€โ”€ api_helper.py       # API ่ฐƒ็”จๅฐ่ฃ…
โ”‚   โ”‚   โ””โ”€โ”€ result.py           # ่ฟ”ๅ›žๆ ผๅผ็ปŸไธ€
โ”‚   โ”‚
โ”‚   โ”œโ”€โ”€ notebook_module/        # Notebook & ้•œๅƒ็ฎก็†
โ”‚   โ”œโ”€โ”€ pool_module/            # ่ต„ๆบๆฑ ็ฎก็†
โ”‚   โ”œโ”€โ”€ node_pool_module/       # ่Š‚็‚นๆฑ ็ฎก็†
โ”‚   โ”œโ”€โ”€ train_module/           # ่ฎญ็ปƒไปปๅŠก็ฎก็†
โ”‚   โ”œโ”€โ”€ infer_v1_module/        # ๆ—ง็‰ˆๆŽจ็†ๆœๅŠก็ฎก็†
โ”‚   โ”œโ”€โ”€ infer_v2_module/        # ๆ–ฐ็‰ˆๆŽจ็†ๆœๅŠก็ฎก็†
โ”‚   โ”œโ”€โ”€ management_module/      # ๅนณๅฐ็ฎก็†: ๅทฅไฝœ็ฉบ้—ด/ๆŽˆๆƒ/้…้ข/ไบ‹ไปถ
โ”‚   โ”œโ”€โ”€ swr_module/             # SWR ้•œๅƒไป“ๅบ“
โ”‚   โ”œโ”€โ”€ vpc_module/             # VPC ่™šๆ‹Ÿ็งๆœ‰ไบ‘
โ”‚   โ”œโ”€โ”€ kms_module/             # KMS ๅฏ†้’ฅ็ฎก็†ๆœๅŠก
โ”‚   โ”œโ”€โ”€ liteserver_module/      # Lite Server ่ฝป้‡่ฎก็ฎ—่Š‚็‚น็ฎก็† (ๅซ HPS/่ถ…่Š‚็‚นๅฎŒๆ•ดๆ”ฏๆŒ)
โ”‚   โ””โ”€โ”€ __pycache__/
โ”‚
โ””โ”€โ”€ references/
    โ”œโ”€โ”€ notebook.md
    โ”œโ”€โ”€ pool.md
    โ”œโ”€โ”€ node_pool.md
    โ”œโ”€โ”€ train.md
    โ”œโ”€โ”€ infer_v1.md
    โ”œโ”€โ”€ infer_v2.md
    โ”œโ”€โ”€ management.md
    โ”œโ”€โ”€ swr.md
    โ”œโ”€โ”€ vpc.md
    โ”œโ”€โ”€ kms.md
    โ””โ”€โ”€ common.md

โœ… Design Features

  1. Modular Design - One file per function, clean structure, easy to maintain
  2. OBS as Primary Storage - Notebook can use OBS buckets directly as root filesystem
  3. OBS Bucket Management - Built-in create/delete/list OBS bucket functions
  4. Environment Variable Auth - Zero configuration for local/CI environments
  5. Unified API Pattern - All functions use consistent calling pattern
  6. Clean Deliverables - No temp scripts, no debug code, no redundant files
  7. Node Pool Management - Independent node pool lifecycle management
  8. Plugin Management - Resource pool plugin listing and creation (ListPoolPlugins, CreatePoolPlugin)
  9. Node Configuration Template Query - Query node configuration template details (ShowNodeConfigTemplate)
  10. Common Infrastructure - Extracted shared infrastructure, improved code reuse
  11. Training Job Management - Full lifecycle: create/list/stop/delete training jobs
  12. Inference Service Management - Old v1 API: create/list/start/stop/update/delete inference services
  13. Workspace & Authorization - Workspace CRUD, auth mode, IAM authorization, quotas, scheduled events
  14. SWR Image Repository - Query training images from SWR
  15. VPC Virtual Private Cloud - Query VPCs, Subnets, Security Groups from Huawei Cloud VPC service
  16. KMS Key Management Service - Query SSH keypairs from Huawei Cloud KMS
  17. Lite Server Management - Full lifecycle management of lightweight compute nodes (51 APIs covering Chapter 22)
  18. Hyper Cluster Management - Create, list, delete hyper clusters for high-performance computing
  19. Hyperinstance Operations - Scale, start, stop, and manage hyper instances
  20. Batch Operations - Bulk actions on multiple Lite Server instances (start/stop/reboot/changeOS/reinstallOS/delete)
  21. Job Management - Create, list, delete jobs on Lite Server instances

๐ŸŽฏ Quick Reference by Use Case

What You Want To DoGo To Module
Start/Stop/Create/Delete Notebooknotebook_module
Save Notebook to Imagenotebook_module
Attach/Detach OBS Storagenotebook_module
Create/Delete OBS Bucketsnotebook_module
Create/Delete Dedicated Resource Poolpool_module
Batch Node Operations (Reboot/Delete/Lock/Resize)pool_module
Scale Node Pool Sizenode_pool_module
Create Additional Node Pools (different flavors)node_pool_module
List Resource Pool Plugins (ListPoolPlugins)node_pool_module
Create Plugin in Resource Pool (CreatePoolPlugin)node_pool_module
Query Node Configuration Template (ShowNodeConfigTemplate)node_pool_module
Create/Stop/Delete Training Jobtrain_module
List Training Jobs (running/pending/completed)train_module
Create/Start/Stop/Delete Inference Serviceinfer_v1_module
List/Get Inference Service Detailsinfer_v1_module
List Models (custom/subscription)infer_v1_module
Create/Start/Stop/Delete Inference Service (ๆ–ฐ็‰ˆ)infer_v2_module
List/Get Inference Service Details (ๆ–ฐ็‰ˆ)infer_v2_module
Query Dedicated Pools & Flavors (ๆ–ฐ็‰ˆ)infer_v2_module
Get Service Exec Login Info (ๆ–ฐ็‰ˆ)infer_v2_module
Batch Delete Inference Services (ๆ–ฐ็‰ˆ)infer_v2_module
Service Version Management (ๆ–ฐ็‰ˆ)infer_v2_module
Service Events & Health Monitoring (ๆ–ฐ็‰ˆ)infer_v2_module
Region Detection & Consistency (ๆ–ฐ็‰ˆ)infer_v2_module
Create/Update/Delete Workspacemanagement_module
Manage Authorization (add/delete/list)management_module
Query/Update Auth Modemanagement_module
List Quotas & Scheduled Eventsmanagement_module
Query SWR Training Imagesswr_module
Query VPC/Subnet/SecurityGroupvpc_module
Query SSH Keypairskms_module
Create/Start/Stop Lite Server Instancesliteserver_module
Manage Hyper Clusters & Hyperinstancesliteserver_module
Batch Operations on Lite Serversliteserver_module
Attach/Detach Volumes & EIPsliteserver_module
Query Flavors & Images for Lite Serverliteserver_module
Create/Manage Jobs on Lite Serverliteserver_module
Infrastructure / Internal Toolscommon_module

What ships with it: 231 files

512.1 KB alongside SKILL.md, 218 of them executable

references/

191 more files not listed here. See all 231 in the repository.

Keep looking

Skills are one crate of 326,506. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.