Huawei Cloud Monitoring Query Skill: Alarm and EPS Lookups

Lets an agent look up Huawei Cloud alarm rules, alarm history, dashboards and enterprise projects through read-only scripts.

Installs: 17.5K; on skills.sh
Stars: 59; on GitHub
Files: 38; besides SKILL.md
Updated: Oct 2026; last repo commit

This is Huawei Cloud's own query skill for Cloud Eye (CES) and Enterprise Project (EPS) management data. The agent runs a bundled Python script per lookup: alarm rules, alarm history, alarm templates, dashboards, notification masks, resource groups, one-click alarms and enterprise projects. SKILL.md says it has no write operations.

It does not read metric values. SKILL.md says it cannot fetch data points such as RDS CPU use over the last hour, and that scripts exist only for CES and EPS. Before the first query the agent runs an environment check that builds a .venv, installs the SDK with pip and tests your keys with an IAM call. SKILL.md writes every script call as skill action=exec with skill:// paths, and the README does not say which agent supports that form.

Good for

Ops teams checking which alarms are set up and which have fired
Admins reviewing dashboards, notification masks and resource groups
Finance and platform teams listing enterprise projects and what is bound to them

Install huawei-cloud-monitoring-query

Claude Code

Copy this prompt.

Install the huawei-cloud-monitoring-query skill in Claude Code. Run this command:

npx skills add huaweicloud/huaweicloud-skills --skill huawei-cloud-monitoring-query

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

There is no separate install step for the Python side. SKILL.md has the agent run scripts/check_env.sh (scripts/check_env.ps1 on Windows) first, which creates a .venv in the skill folder and pip-installs the Huawei Cloud SDK packages. Set the key variables in your shell before you start the agent. SKILL.md writes every script call as skill action=exec with skill:// paths, and the README does not say which agent supports that form. The repo has no plugin manifest, so there is no plugin route, and there is no Claude app tab because the skill runs local Python.

Built from the package data we hold. Not yet tested by us. Source.

What's in the skill

When it loads

Queries Huawei Cloud monitoring and enterprise project resources (CES/EPS). Covers alarm rules, alarm histories, alarm templates, dashboards, notification masks, resource groups, one-click alarms, and enterprise projects (list/detail/quotas/bound resources/migration records). Only management-plane queries are supported; querying raw metric data points (e.g., ShowMetricData/ListMetrics) and service-level metric values (RDS, DCS, DMS, EVS, ELB, etc.) is NOT supported. No write operations. Use this skill when the user needs to check alarm status, view monitoring dashboards, query alarm rules, or manage enterprise project info. Triggers: 云监控, 告警, 告警规则, 告警历史, 仪表盘, 企业项目, CES, EPS, 告警模板, 资源分组, alarm, monitoring, alert.

Sections in SKILL.md

Overview
Capability Scope
Out of Scope (Not Supported)
Usage Principles
Prerequisites
Execution Flow
Step 1: Environment Preparation
Step 2: Identify and Execute Query Scripts
Directory Structure
Parameter Confirmation
Output Format
Verification Methods
Best Practices
Reference Documentation
Notes

Key facts

Author: Huawei Cloud (the skill is in the vendor's own huaweicloud/huaweicloud-skills repo; the README calls these officially maintained Huawei Cloud Agent Skills)
License: MIT (LICENSE file and README badge; the LICENSE copyright line still reads "[year] [fullname]")
Installs on skills.sh: 17,523 (Nov 2, 2026)
skills.sh audits: Gen Agent Trust Hub pass, Socket warn, Snyk warn (Nov 2, 2026)
GitHub stars: 59 for the whole huaweicloud/huaweicloud-skills repo (Nov 2, 2026)
Last repo commit: Oct 31, 2026
What it includes: 39 files including SKILL.md (counted from the repo tree): 33 Python files, 3 Markdown files, check_env.sh, check_env.ps1 and requirements.txt. The Python files are 30 query scripts in 3 folders (ces 19, eps 11, iam 1) plus config.py and ensure_env.py.
Needs: Python 3.6 or newer, plus HW_ACCESS_KEY and HW_SECRET_KEY in the environment (SKILL.md)
Works in: The README names no client; it says the skills are "compatible with mainstream AI Agents"

What it includes

Builds the Python environment and tests your keys before any query
Runs scripts/check_env.sh (or check_env.ps1 on Windows). It checks Python, installs the SDK packages into a .venv, imports them and makes an IAM call to confirm the key works.
Try: “Check my Huawei Cloud setup, then show my alarm rules.”
Runs code · Environment check
19 scripts for alarm rules, history, templates and dashboards
Reads Cloud Eye alarm rules, alarm histories, alarm templates, dashboards, notification masks, resource groups and one-click alarms.
Try: “Which alarm rules are set up, and which fired recently?”
Network · CES queries
11 scripts for enterprise projects and what is bound to them
Lists enterprise projects, their details and quotas, bound resources and migration records.
Try: “List my enterprise projects and the resources in each.”
Network · EPS queries
Finds the project ID for a region
One script in scripts/iam asks IAM for the project that matches a region.
Network · Project ID lookup
Tells you what it will not look up
SKILL.md names metric data points (ShowMetricData, ListMetrics), service metric values for RDS, DCS, DMS, EVS, ELB and ECS, and any service other than CES and EPS as unsupported, and says to state that plainly.
Read · Out-of-scope list

From the official docs

Permissions

Medium risk

Your cloud key is used by scripts that run on your machine and talk to Huawei Cloud, and TLS certificate checks are turned off. Use a key that can only read, on a network you trust.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It turns off TLS certificate checks
scripts/config.py sets ignore_ssl_verification = True for the Huawei Cloud SDK, and scripts/ensure_env.py switches Python to an unverified HTTPS context. Use it only on a network you trust. github.com
Its setup script installs packages with pip
ensure_env.py creates a .venv in the skill folder and runs pip install -r requirements.txt, trying repo.huaweicloud.com, pypi.tuna.tsinghua.edu.cn and mirrors.aliyun.com with --trusted-host before plain PyPI. If pip is missing it can fetch get-pip.py from mirrors.huaweicloud.com or bootstrap.pypa.io. github.com
It uses your cloud keys and calls IAM to test them
The scripts read the HW_ACCESS_KEY and HW_SECRET_KEY environment variables. The environment check makes a live IAM call to confirm they work. Use a key that can only read. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	SKILL.md says this is a read-only query skill. There is no switch to enforce it.
Talks to	Huawei Cloud API endpoints through the Huawei Cloud Python SDK. repo.huaweicloud.com, pypi.tuna.tsinghua.edu.cn, mirrors.aliyun.com and PyPI during the pip install. mirrors.huaweicloud.com and bootstrap.pypa.io if pip must be bootstrapped
Sign-in and keys	The scripts read the key from environment variables. SKILL.md says not to print the values.
Scopes	Reads HW_ACCESS_KEY, HW_SECRET_KEY, HW_REGION_NAME, HW_PROJECT_ID and HW_SECURITY_TOKEN from your environment Calls Huawei Cloud APIs to look up alarm rules, dashboards and enterprise projects Writes a .venv folder inside the skill folder

Questions

What does huawei-cloud-monitoring-query do?

It makes an agent look up Huawei Cloud Cloud Eye alarm data and Enterprise Project data by running bundled read-only Python scripts. It does not read raw metric values. SKILL.md says it only performs queries and does no write operations.

Does the monitoring query skill need an API key?

Yes. SKILL.md lists HW_ACCESS_KEY and HW_SECRET_KEY as required, HW_REGION_NAME as optional with a default of cn-north-4, and HW_SECURITY_TOKEN for temporary keys. Our advice: use an IAM user that can only read, since the skill only queries.

Is huawei-cloud-monitoring-query official?

It is in Huawei Cloud's own huaweicloud/huaweicloud-skills repo, which the README calls officially maintained Huawei Cloud Agent Skills.

Is the monitoring query skill safe to run?

It runs Python scripts on your machine with your cloud key, and SKILL.md says it only queries. Its setup also installs packages with pip and turns off TLS certificate checks. See the safety box for the details, and use a read-only key.

Does the monitoring query skill work in Claude Code, Codex or the Claude app?

The install box has Claude Code, Codex and standalone tabs. Huawei says only that the skills are compatible with mainstream AI Agents and names none. SKILL.md tells the agent to run scripts with a skill action=exec call and skill:// paths, and Huawei does not say which agents support that call, so check that yours does. There is no Claude app tab because the skill runs local Python.

Can it show CPU or disk graphs?

No. SKILL.md lists raw metric data points and service-level metric values (RDS, DCS, DMS, EVS, ELB, ECS) as unsupported. For those it points to the Huawei Cloud console or the CES API. The OBS stats skill queries a few OBS metrics through the hcloud CLI.

Information

updated Nov 2, 2026

Repository: huaweicloud/huaweicloud-skills
Folder: skills/monitoring/ops/huawei-cloud-monitoring-query

huaweicloud/huaweicloud-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 6

huawei-cloud-monitoring-query SKILL.md · github.com
huaweicloud-skills README · github.com
huaweicloud-skills LICENSE · github.com
huawei-cloud-monitoring-query ensure_env.py · github.com
huawei-cloud-monitoring-query config.py · github.com
huawei-cloud-monitoring-query requirements.txt · github.com