huawei-cloud-modelarts-resource-pool-management

Installation
SKILL.md

Huawei Cloud ModelArts Resource Pool Management

Full lifecycle management for ModelArts dedicated resource pools and node pools via hcloud CLI — 53 operations across 10 functional domains.


Overview

This skill enables users to manage Huawei Cloud ModelArts dedicated resource pools (专属资源池) and node pools through the hcloud CLI. It covers the complete lifecycle including resource pool CRUD, node management, node pool management, network resources, tag management, plugin management, workloads, scheduled events, OS configuration, and resource flavor queries. Pricing inquiry for chargeable operations is included via scripts/query-pricing.sh (KooCLI 7.3.12 no longer supports BSS, so the script fails loudly and a console/API Explorer fallback is documented).

⚠️ Scope boundary: This skill is used ONLY for ModelArts dedicated resource pool / node pool / network / tag / plugin management. It does NOT handle OBS buckets, ECS instances, or any other Huawei Cloud service operations. For non-ModelArts requests (e.g., "query my OBS bucket"), decline and point the user to the appropriate service-specific skill.

Known Capability Limitations (verified by real testing)

  • CreatePool flavors must carry the .d suffix: Only flavors like modelarts.vm.cpu.16u64g.d can create a dedicated pool. Flavors WITHOUT the .d suffix (e.g., modelarts.vm.cpu.8ud) fail with ModelArts.50004000 Invalid flavor ... is not available for your project. Always verify the targeted flavor with ListResourceFlavors before CreatePool.
  • BatchResetPoolNodes is NOT available for regular VM pools: rollingConfig.strategy has no valid value for VM pools, so batch reset only works when the pool supports a valid strategy. For regular VM nodes, resets must be done by deleting and recreating the nodes instead.
  • BatchResizePoolNodes only supports hyperinstance nodes: resizing is not supported for regular VM nodes; use delete-recreate as the fallback for VM pools.

Architecture

Installs
68
GitHub Stars
59
First Seen
Sep 7, 2026