huaweicloud-devkit 1.0.2-next.3 → 1.0.3-dev.0

package/plugins/huaweicloud-core/skills/huawei-apig/SKILL.md

@@ -1,164 +1,164 @@
----
-name: huawei-apig
-description: "Use when creating or managing API Gateway (APIG). Covers API creation, throttling, auth (IAM/APP/basic), CORS, publishing, instance management. Triggers: APIG, API gateway, throttling, publish API. NOT for: FunctionGraph triggers (use huawei-functiongraph)."
-version: 1
----
-
-# Huawei Cloud APIG
-
-**STOP - Do not answer from general knowledge.** Follow the procedure below.
-
-## Overview
-
-Domain expertise for Huawei Cloud API Gateway (APIG). Covers instance lifecycle, API group/API creation, publishing, and FunctionGraph trigger integration.
-
-Always discover parameters with `hcloud APIG <Operation> --help` before executing.
-
-## Critical Warnings
-
-| Trap | Why |
-|------|-----|
-| API group region-locked | Cannot move across regions |
-| Throttling per-API default | Use app-level quotas for per-user limits |
-| CORS must be explicit | OPTIONS preflight fails until configured |
-| `BASIC` spec has no public IP | Use `PROFESSIONAL` + `elb` provider for public access |
-| Instance creation takes 5-15min | Long-running async operation. State is **Running** (NOT "SUCCESS"). Poll with `ListInstancesV2`, wait for `status == "Running"` |
-| `sl_domain` is from API **Group** | NOT from Instance. Get it from `CreateApiGroupV2` or `ListApiGroupsV2` response |
-| API name must NOT have hyphens | `[a-zA-Z0-9_]+` only. Hyphens cause regex validation failure |
-| VPC params need prefix | `--vpc.name=<n>` / `--subnet.vpc_id=<id>` / `--security_group.name=<n>` with KooCLI 7.x |
-
-## Instance Management
-
-### Check Existing Instances
-
-```bash
-hcloud APIG ListInstancesV2 --cli-region=<r>
-```
-
-### Create Instance
-
-```bash
-hcloud APIG CreateInstanceV2 --help
-```
-
-Key gotchas when creating:
-
-| Param | Note |
-|-------|------|
-| `--spec_id` | `BASIC` (no public access), `PROFESSIONAL` (requires `--loadbalancer_provider`) |
-| `--loadbalancer_provider` | `elb` for public access (supports `AddIngressEipV2`), `lvs` for internal only (supports `AddEipV2`) |
-| `--enterprise_project_id` | **Required** for enterprise accounts. Use `"0"` for default project |
-| `--available_zone_ids` | Use AZ code like `ap-southeast-3a`, NOT UUID from `ListAvailableZonesV2` |
-| `--vpc_id`, `--subnet_id` | Must exist in the target region |
-| `--security_group_id` | **Required**. Create a security group first via `VPC CreateSecurityGroup` (VPC v3 API — no `vpc_id` param needed) |
-
-### Add Public Access (ELB Provider Only)
-
-`BASIC` instances have no public IP. Use `PROFESSIONAL` + `elb` provider for public access (`lvs` is internal-only).
-
-```bash
-hcloud APIG AddIngressEipV2 \
-  --instance_id=<id> \
-  --bandwidth_charging_mode=bandwidth \
-  --bandwidth_size=<size>
-```
-
-To use an existing EIP instead of creating a new one:
-
-```bash
-hcloud APIG AddIngressEipV2 \
-  --instance_id=<id> \
-  --eip_id=<existing-eip-id>
-```
-
-> `AddIngressEipV2` only works with `elb` provider. `AddEipV2` (without "Ingress") requires `lvs` provider. Bandwidth minimum is 5 Mbps.
-
-**Verification** — After binding, poll the instance to confirm EIP is active:
-
-```bash
-hcloud APIG ListInstancesV2 --cli-region=<region> --instance_id=<id> --cli-output=json | jq '.instances[0].eip_address'
-```
-
-Wait for `eip_address` to show a valid IP (may take up to 2 minutes). If `eip_address` is still `null` after 2 minutes, the EIP may not have been assigned.
-
-> **Trap**: `sl_domain` from `CreateApiGroupV2` is an **internal-only** domain (e.g., `*.apic.cn-north-4.huaweicloudapis.com`). It may resolve to internal IP only (NXDOMAIN from public internet). **For public access, always use the `eip_address` from the instance**, not the `sl_domain` or trigger `invoke_url`.
-
-## API Group
-
-```bash
-hcloud APIG CreateApiGroupV2 --instance_id=<id> --name=<n>
-```
-
-| Param | Note |
-|-------|------|
-| `--name` | Group name (required) |
-| `--remark` | Description (optional) |
-
-## API Management
-
-```bash
-hcloud APIG CreateApiV2 --help
-```
-
-## Publishing
-
-```bash
-hcloud APIG BatchPublishOrOfflineApiV2 \
-  --instance_id=<id> \
-  --action=online \
-  --env_id=<env-id> \
-  --apis.1=<api-id>
-```
-
-> The operation is `BatchPublishOrOfflineApiV2`, NOT `PublishApiV2`. The parameter is `apis.1` (1-based array), NOT `api_ids`. For multiple APIs, use `--apis.1`, `--apis.2`, etc.
-
-## Throttling
-
-```bash
-hcloud APIG CreateThrottlingPolicyV2 --name=<n> --api_call_limits=1000
-```
-
-## Common Workflows
-
-| Task | Operation |
-|------|-----------|
-| List instances | `ListInstancesV2` |
-| Create instance | `CreateInstanceV2` |
-| Delete instance | `DeleteInstancesV2` |
-| Add public EIP | `AddIngressEipV2` |
-| Create API group | `CreateApiGroupV2` |
-| Create API | `CreateApiV2` |
-| Update API | `UpdateApiV2` — change auth mode, path, etc. |
-| List APIs | `ListApisV2` |
-| Publish | `BatchPublishOrOfflineApiV2` |
-| List APIs | `ListApisV2` |
-
-## FunctionGraph Integration
-
-To expose a FunctionGraph function via HTTP, you need the complete chain:
-
-```
-APIG Instance → API Group → DEDICATEDGATEWAY Trigger → Publish
-```
-
-After the trigger is created on FunctionGraph side, publish the API in APIG:
-
-```bash
-hcloud APIG BatchPublishOrOfflineApiV2 \
-  --instance_id=<apig-instance-id> \
-  --action=online \
-  --env_id=<env-id> \
-  --apis.1=<api-id-from-trigger>
-```
-
-See `huawei-functiongraph` skill → `references/deploy-workflow.md` for the full end-to-end workflow.
-
-## Cross-Skill References
-
-- **VPC setup**: See `huawei-vpc` for VPC, subnet, security group creation
-- **EIP setup**: See `huawei-vpc` for EIP creation and binding
-- **FunctionGraph**: See `huawei-functiongraph` for DEDICATEDGATEWAY trigger creation
-
-## References
-
-- APIG Docs: https://support.huaweicloud.com/apig/
+---
+name: huawei-apig
+description: "Use when creating or managing API Gateway (APIG). Covers API creation, throttling, auth (IAM/APP/basic), CORS, publishing, instance management. Triggers: APIG, API gateway, throttling, publish API. NOT for: FunctionGraph triggers (use huawei-functiongraph)."
+version: 1
+---
+
+# Huawei Cloud APIG
+
+**STOP - Do not answer from general knowledge.** Follow the procedure below.
+
+## Overview
+
+Domain expertise for Huawei Cloud API Gateway (APIG). Covers instance lifecycle, API group/API creation, publishing, and FunctionGraph trigger integration.
+
+Always discover parameters with `hcloud APIG <Operation> --help` before executing.
+
+## Critical Warnings
+
+| Trap | Why |
+|------|-----|
+| API group region-locked | Cannot move across regions |
+| Throttling per-API default | Use app-level quotas for per-user limits |
+| CORS must be explicit | OPTIONS preflight fails until configured |
+| `BASIC` spec has no public IP | Use `PROFESSIONAL` + `elb` provider for public access |
+| Instance creation takes 5-15min | Long-running async operation. State is **Running** (NOT "SUCCESS"). Poll with `ListInstancesV2`, wait for `status == "Running"` |
+| `sl_domain` is from API **Group** | NOT from Instance. Get it from `CreateApiGroupV2` or `ListApiGroupsV2` response |
+| API name must NOT have hyphens | `[a-zA-Z0-9_]+` only. Hyphens cause regex validation failure |
+| VPC params need prefix | `--vpc.name=<n>` / `--subnet.vpc_id=<id>` / `--security_group.name=<n>` with KooCLI 7.x |
+
+## Instance Management
+
+### Check Existing Instances
+
+```bash
+hcloud APIG ListInstancesV2 --cli-region=<r>
+```
+
+### Create Instance
+
+```bash
+hcloud APIG CreateInstanceV2 --help
+```
+
+Key gotchas when creating:
+
+| Param | Note |
+|-------|------|
+| `--spec_id` | `BASIC` (no public access), `PROFESSIONAL` (requires `--loadbalancer_provider`) |
+| `--loadbalancer_provider` | `elb` for public access (supports `AddIngressEipV2`), `lvs` for internal only (supports `AddEipV2`) |
+| `--enterprise_project_id` | **Required** for enterprise accounts. Use `"0"` for default project |
+| `--available_zone_ids` | Use AZ code like `ap-southeast-3a`, NOT UUID from `ListAvailableZonesV2` |
+| `--vpc_id`, `--subnet_id` | Must exist in the target region |
+| `--security_group_id` | **Required**. Create a security group first via `VPC CreateSecurityGroup` (VPC v3 API — no `vpc_id` param needed) |
+
+### Add Public Access (ELB Provider Only)
+
+`BASIC` instances have no public IP. Use `PROFESSIONAL` + `elb` provider for public access (`lvs` is internal-only).
+
+```bash
+hcloud APIG AddIngressEipV2 \
+  --instance_id=<id> \
+  --bandwidth_charging_mode=bandwidth \
+  --bandwidth_size=<size>
+```
+
+To use an existing EIP instead of creating a new one:
+
+```bash
+hcloud APIG AddIngressEipV2 \
+  --instance_id=<id> \
+  --eip_id=<existing-eip-id>
+```
+
+> `AddIngressEipV2` only works with `elb` provider. `AddEipV2` (without "Ingress") requires `lvs` provider. Bandwidth minimum is 5 Mbps.
+
+**Verification** — After binding, poll the instance to confirm EIP is active:
+
+```bash
+hcloud APIG ListInstancesV2 --cli-region=<region> --instance_id=<id> --cli-output=json | jq '.instances[0].eip_address'
+```
+
+Wait for `eip_address` to show a valid IP (may take up to 2 minutes). If `eip_address` is still `null` after 2 minutes, the EIP may not have been assigned.
+
+> **Trap**: `sl_domain` from `CreateApiGroupV2` is an **internal-only** domain (e.g., `*.apic.cn-north-4.huaweicloudapis.com`). It may resolve to internal IP only (NXDOMAIN from public internet). **For public access, always use the `eip_address` from the instance**, not the `sl_domain` or trigger `invoke_url`.
+
+## API Group
+
+```bash
+hcloud APIG CreateApiGroupV2 --instance_id=<id> --name=<n>
+```
+
+| Param | Note |
+|-------|------|
+| `--name` | Group name (required) |
+| `--remark` | Description (optional) |
+
+## API Management
+
+```bash
+hcloud APIG CreateApiV2 --help
+```
+
+## Publishing
+
+```bash
+hcloud APIG BatchPublishOrOfflineApiV2 \
+  --instance_id=<id> \
+  --action=online \
+  --env_id=<env-id> \
+  --apis.1=<api-id>
+```
+
+> The operation is `BatchPublishOrOfflineApiV2`, NOT `PublishApiV2`. The parameter is `apis.1` (1-based array), NOT `api_ids`. For multiple APIs, use `--apis.1`, `--apis.2`, etc.
+
+## Throttling
+
+```bash
+hcloud APIG CreateThrottlingPolicyV2 --name=<n> --api_call_limits=1000
+```
+
+## Common Workflows
+
+| Task | Operation |
+|------|-----------|
+| List instances | `ListInstancesV2` |
+| Create instance | `CreateInstanceV2` |
+| Delete instance | `DeleteInstancesV2` |
+| Add public EIP | `AddIngressEipV2` |
+| Create API group | `CreateApiGroupV2` |
+| Create API | `CreateApiV2` |
+| Update API | `UpdateApiV2` — change auth mode, path, etc. |
+| List APIs | `ListApisV2` |
+| Publish | `BatchPublishOrOfflineApiV2` |
+| List APIs | `ListApisV2` |
+
+## FunctionGraph Integration
+
+To expose a FunctionGraph function via HTTP, you need the complete chain:
+
+```
+APIG Instance → API Group → DEDICATEDGATEWAY Trigger → Publish
+```
+
+After the trigger is created on FunctionGraph side, publish the API in APIG:
+
+```bash
+hcloud APIG BatchPublishOrOfflineApiV2 \
+  --instance_id=<apig-instance-id> \
+  --action=online \
+  --env_id=<env-id> \
+  --apis.1=<api-id-from-trigger>
+```
+
+See `huawei-functiongraph` skill → `references/deploy-workflow.md` for the full end-to-end workflow.
+
+## Cross-Skill References
+
+- **VPC setup**: See `huawei-vpc` for VPC, subnet, security group creation
+- **EIP setup**: See `huawei-vpc` for EIP creation and binding
+- **FunctionGraph**: See `huawei-functiongraph` for DEDICATEDGATEWAY trigger creation
+
+## References
+
+- APIG Docs: https://support.huaweicloud.com/apig/

package/plugins/huaweicloud-core/skills/huawei-billing/SKILL.md

@@ -1,48 +1,48 @@
----
-name: huawei-billing
-description: "Use when querying bills, costs, resource usage, or billing details on Huawei Cloud (BSS). Triggers: billing, BSS, cost, bill, expense, usage report, resource usage, budget. NOT for: resource management (use huawei-ecs etc.), creating resources."
-version: 1
----
-
-# Huawei Cloud Billing (BSS)
-
-**STOP - Do not answer from general knowledge.** Follow the procedure below.
-
-Always run `hcloud BSS <Operation> --help` before constructing commands to discover exact parameter names and requirements.
-
-## Overview
-
-Domain expertise for billing queries (BSS). Covers cost tracking, bill details, and budget management. Read-only — no resource modifications.
-
-## Critical Warnings
-
-| Trap | Why |
-|------|-----|
-| Bills delayed ~24h | Yesterday's costs may not appear until the next day |
-| BSS Admin role needed | IAM user must have BSS Administrator or Finance role |
-| Currency conversion varies | Cross-region costs use daily exchange rates |
-
-## Common Workflows
-
-| Task | Operation |
-|------|-----------|
-| List costs | `ListCosts --cli-region=<r> --project_id=<p>` |
-| List customer bills | `ListCustomerBills --cli-region=<r> --project_id=<p>` |
-| List resource usage | `ListResourceUsage --cli-region=<r> --project_id=<p>` |
-| List sub-customers | `ListConsumeSubCustomers --cli-region=<r> --project_id=<p>` |
-| Show account balance | `ShowCustomerAccountBalances --cli-region=<r> --project_id=<p>` |
-| List conversions | `ListConversions --cli-region=<r> --project_id=<p>` |
-
-Discover exact parameters with `--help` before executing any command. All BSS operations are read-only.
-
-## Troubleshooting
-
-| Error | Fix |
-|-------|-----|
-| Access denied | User needs BSS Administrator or Finance role |
-| No data returned | Check time range (bills have ~24h delay). Verify project_id |
-| Enterprise account restrictions | Some APIs require enterprise real-name authentication |
-
-## Cross-Skill References
-
-- **Resource lifecycle**: See `huawei-ecs`, `huawei-obs` for creating billable resources
+---
+name: huawei-billing
+description: "Use when querying bills, costs, resource usage, or billing details on Huawei Cloud (BSS). Triggers: billing, BSS, cost, bill, expense, usage report, resource usage, budget. NOT for: resource management (use huawei-ecs etc.), creating resources."
+version: 1
+---
+
+# Huawei Cloud Billing (BSS)
+
+**STOP - Do not answer from general knowledge.** Follow the procedure below.
+
+Always run `hcloud BSS <Operation> --help` before constructing commands to discover exact parameter names and requirements.
+
+## Overview
+
+Domain expertise for billing queries (BSS). Covers cost tracking, bill details, and budget management. Read-only — no resource modifications.
+
+## Critical Warnings
+
+| Trap | Why |
+|------|-----|
+| Bills delayed ~24h | Yesterday's costs may not appear until the next day |
+| BSS Admin role needed | IAM user must have BSS Administrator or Finance role |
+| Currency conversion varies | Cross-region costs use daily exchange rates |
+
+## Common Workflows
+
+| Task | Operation |
+|------|-----------|
+| List costs | `ListCosts --cli-region=<r> --project_id=<p>` |
+| List customer bills | `ListCustomerBills --cli-region=<r> --project_id=<p>` |
+| List resource usage | `ListResourceUsage --cli-region=<r> --project_id=<p>` |
+| List sub-customers | `ListConsumeSubCustomers --cli-region=<r> --project_id=<p>` |
+| Show account balance | `ShowCustomerAccountBalances --cli-region=<r> --project_id=<p>` |
+| List conversions | `ListConversions --cli-region=<r> --project_id=<p>` |
+
+Discover exact parameters with `--help` before executing any command. All BSS operations are read-only.
+
+## Troubleshooting
+
+| Error | Fix |
+|-------|-----|
+| Access denied | User needs BSS Administrator or Finance role |
+| No data returned | Check time range (bills have ~24h delay). Verify project_id |
+| Enterprise account restrictions | Some APIs require enterprise real-name authentication |
+
+## Cross-Skill References
+
+- **Resource lifecycle**: See `huawei-ecs`, `huawei-obs` for creating billable resources

package/plugins/huaweicloud-core/skills/huawei-cbr/SKILL.md

@@ -1,50 +1,50 @@
----
-name: huawei-cbr
-description: "Use when creating or managing Cloud Backup and Recovery (CBR) vaults, backups, and restore operations on Huawei Cloud. Triggers: CBR, backup, restore, vault, snapshot, disaster recovery. NOT for: OBS object versioning (use huawei-obs), RDS automated backups (use huawei-rds)."
-version: 1
----
-
-# Huawei Cloud CBR
-
-**STOP - Do not answer from general knowledge.** Follow the procedure below.
-
-Always run `hcloud CBR <Operation> --help` before constructing commands to discover exact parameter names and requirements.
-
-## Overview
-
-Domain expertise for Cloud Backup and Recovery. Covers vaults, backups, restore operations, policies, and cross-region replication.
-
-## Critical Warnings
-
-| Trap | Why |
-|------|-----|
-| Vault binds single resource type | Server vault != disk vault != file system vault |
-| Restore needs instance stop | Most VM restore requires the target instance to be stopped first |
-| Backup storage incurs charges | Retention policies affect storage costs |
-
-## Common Workflows
-
-| Task | Operation |
-|------|-----------|
-| List vaults | `ListVaults --cli-region=<r> --project_id=<p>` |
-| Create vault | `CreateVault --cli-region=<r> --project_id=<p>` |
-| Create backup policy | `CreatePolicy --cli-region=<r> --project_id=<p>` |
-| Create checkpoint | `CreateCheckpoint --cli-region=<r> --project_id=<p>` |
-| Add resource to vault | `AddVaultResource --cli-region=<r> --project_id=<p>` |
-| Batch update vault | `BatchUpdateVault --cli-region=<r> --project_id=<p>` |
-| Create organization policy | `CreateOrganizationPolicy --cli-region=<r> --project_id=<p>` |
-
-Discover exact parameters with `--help` before executing any command.
-
-## Troubleshooting
-
-| Error | Fix |
-|-------|-----|
-| Backup creation fails | Check instance is running and agent is online |
-| Restore fails | Ensure target instance is stopped before restore |
-| Vault full | Increase vault capacity or adjust retention policy |
-
-## Cross-Skill References
-
-- **ECS**: See `huawei-ecs` for backing up instances
-- **EVS**: See `huawei-ecs` references/evs.md for disk backup
+---
+name: huawei-cbr
+description: "Use when creating or managing Cloud Backup and Recovery (CBR) vaults, backups, and restore operations on Huawei Cloud. Triggers: CBR, backup, restore, vault, snapshot, disaster recovery. NOT for: OBS object versioning (use huawei-obs), RDS automated backups (use huawei-rds)."
+version: 1
+---
+
+# Huawei Cloud CBR
+
+**STOP - Do not answer from general knowledge.** Follow the procedure below.
+
+Always run `hcloud CBR <Operation> --help` before constructing commands to discover exact parameter names and requirements.
+
+## Overview
+
+Domain expertise for Cloud Backup and Recovery. Covers vaults, backups, restore operations, policies, and cross-region replication.
+
+## Critical Warnings
+
+| Trap | Why |
+|------|-----|
+| Vault binds single resource type | Server vault != disk vault != file system vault |
+| Restore needs instance stop | Most VM restore requires the target instance to be stopped first |
+| Backup storage incurs charges | Retention policies affect storage costs |
+
+## Common Workflows
+
+| Task | Operation |
+|------|-----------|
+| List vaults | `ListVaults --cli-region=<r> --project_id=<p>` |
+| Create vault | `CreateVault --cli-region=<r> --project_id=<p>` |
+| Create backup policy | `CreatePolicy --cli-region=<r> --project_id=<p>` |
+| Create checkpoint | `CreateCheckpoint --cli-region=<r> --project_id=<p>` |
+| Add resource to vault | `AddVaultResource --cli-region=<r> --project_id=<p>` |
+| Batch update vault | `BatchUpdateVault --cli-region=<r> --project_id=<p>` |
+| Create organization policy | `CreateOrganizationPolicy --cli-region=<r> --project_id=<p>` |
+
+Discover exact parameters with `--help` before executing any command.
+
+## Troubleshooting
+
+| Error | Fix |
+|-------|-----|
+| Backup creation fails | Check instance is running and agent is online |
+| Restore fails | Ensure target instance is stopped before restore |
+| Vault full | Increase vault capacity or adjust retention policy |
+
+## Cross-Skill References
+
+- **ECS**: See `huawei-ecs` for backing up instances
+- **EVS**: See `huawei-ecs` references/evs.md for disk backup