huawei-cloud-billing-scout

Q: How do I install huawei-cloud-billing-scout?

Run npx skills add https://github.com/huaweicloud/huaweicloud-skills --skill huawei-cloud-billing-scout in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/huawei-cloud-billing-scout/ (personal) or .claude/skills/huawei-cloud-billing-scout/ (project).

by huaweicloudDevOps & Cloud

Huawei Cloud BSS billing (not AWS/Azure/other clouds; refuses pricing quotes, real-name review, and any non-billing scope): balance, spend, attribution, reconciliation, coupons, stored-value cards, enterprise/partner billing. One-page briefing via hcloud. Use only when the user explicitly mentions 华为云 / Huawei Cloud / BSS and 余额/账单/对账/资源包/代金券/储值卡/企业或伙伴账务; refuses pay, renew, refund, delete.

Installs: 17.2K
Last 24h: +104
Trending rank: #1930
Security: Pass

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/huaweicloud/huaweicloud-skills --skill huawei-cloud-billing-scout

01
Install
Run the command in your project. The CLI asks which agents to install for.
02
Describe the task
Ask your agent for something this skill covers. It loads huawei-cloud-billing-scout automatically when relevant.
03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.

Security audits

Independent scans of the skill's instructions and bundled files.

Gen Agent Trust Hub PassSAFE ·The skill provides a read-only billing reconciliation tool for Huawei Cloud using the official hcloud CLI. Security analysis confirms that all external downloads and command executions are directed at official Huawei Cloud infrastructure, and destructive commands are appropriately scoped to tool uninstallation.

Socket PassNo alerts

Snyk PassLOW ·Risk: LOW · No issues

SKILL.md

The instructions your agent reads when the skill is triggered.

Huawei Cloud · How Much Spent & Why Charged · Read-Only Reconciliation

Huawei Cloud Read-Only Billing — Spend, Charges & Reconciliation

Using hcloud ≥7.2.2 and BSS read-only IAM, answer in one conversation: how much spent, why charged, where the difference is, what evidence is still missing. Query only, no modifications, no account operations on behalf of users.

Prerequisites

Prerequisite check: Huawei Cloud CLI (hcloud) >= 7.2.2 required Run hcloud version to verify version >= 7.2.2, and hcloud configure list to check profile exists. Run hcloud configure set --cli-lang=cn — BSS requires Chinese mode. If not installed or version is too low, see references/cli-installation-guide.md for installation guide.

hcloud version
hcloud configure list
hcloud configure set --cli-lang=cn

Principles

North Star — Any assertion must be reducible to the quadruple "fact × grain × money_basis × scope/billing_period"; irreducible assertions only list gaps, no conclusions.

Three-Set First — Stop if any of scope, billing_period (time), or money_basis is missing; ask only the point that would change the verification path, no clarification questionnaire.
Single Fact No Mixing — Monthly summary, resource detail, monthly amortization, and orders are each a distinct fact type with different grains; no cross-summing, no substitution.
Evidence Boundary Self-Consistent — Each entity only answers questions within its evidence_boundary; speculation, pending checks, and responsibility judgments must first verify whether evidence falls within the boundary.

Division of Labor

SKILL.md defines behavior; semantic/catalog.yml defines entry points and required context; semantic/billing-ontology.yml defines facts, grains, money_basis, and source_operations; references/related-commands.md defines copyable BSS templates and pagination limits.

Four types of billing questions and routing: see references/semantic/catalog.yml.

Verification Path

Huawei Cloud Gate — Before matching in catalog.yml: if user has not indicated Huawei Cloud / BSS / this skill's billing scope, and cannot be determined from conversation as the Huawei Cloud account for current hcloud profile, first ask one confirmation "Query current configured Huawei Cloud account and billing period?"; without confirmation, no BSS query execution. Non-Huawei Cloud or other cloud vendor billing → only state out of scope, no evidence gathering.

Phase	Task	Reference File	Forbidden
Lock Money Basis	Lock three-set	`catalog.yml` → `required_context`	Ask one at a time if missing, not all together
Select Entry Point	Match `triggers` to get `entry_point`, obtain `ontology_entities`	`catalog.yml` → `entry_points`	No cross-entry-point summing or borrowing
Gather Evidence	Minimal read-only queries within `evidence_boundary`; first query copy from `related-commands.md` current entry `####` template	`billing-ontology.yml` + `related-commands.md`	No `--help` to discover operations; no self-constructed JSON; no full detail list first
Deliver	Conclusion first then facts; conclusion must be reducible to quadruple	This document "Response Requirements"	No sending command process; no transferring investigation burden

Two defaults and one fixed config only valid during evidence gathering phase:

Reconciliation — When user has expressed read-only intent, default to current profile and current (or given) billing period, gather evidence per related-commands.md reconciliation order; ask one ID at a time only when blocking ID is missing.
Enterprise / Partner — When ontology requires customer_id or other prerequisite IDs, first provide read-only acquisition path or one clarification, then make responsibility judgment.
BSS Endpoint — All hcloud BSS calls require --cli-region=cn-north-1; no substitution with profile or other regions. [Fixed config, non-adjustable]

Hard Constraints (Non-Negotiable)

The following three are derived from principles, non-negotiable.

Read-Only

No write operations that change funds, orders, resources, or identity status; refuse payment, refund, unsubscribe, delete, recovery, create, update, send verification code, change balance. Why: Once a write occurs, the fact state that assertions depend on is contaminated by itself. List*/Show* with Change in name are still read-only ledgers, not in this category.

No Leak

No output of credentials, long identifiers that can restore identity, complete business IDs, profile / region. Why: Identity dimension disclosure exceeds evidence_boundary response scope; delivery value is unrelated to disclosure degree.

No Extrapolation

Pagination, partial time windows, sampling, and zero or low-amount results cannot be stated as whole account, all services, final billing, or no subsequent charges. Why: Grain does not allow amplification; conclusions at local grain cannot be declared as facts at coarser grain unless evidence money_basis already covers entire month.

Response Requirements

Briefing-style delivery: conclusion first, facts later; only write what was found, state money_basis clearly.

Like a Brief — Summary one to three sentences, state scope / billing_period / money_basis, answer spend, charge reason, difference, and what's still missing; with evidence then qualitative, without evidence then mark uncertain. Fact points use · list or short paragraphs, no Markdown tables (IM-friendly).
Only Trust Evidence — Points only list content that was found, use business names (consistent with console, no API names); speculation and pending checks only in summary; no amount written for unchecked items.
Delivery Baseline — No sending: raw responses, command text, complete business IDs, credentials, profile / region. Gaps only provide one read-only next step (business wording), not "please reconcile yourself".

Boundaries

Service Scope — Only BSS read-only billing (balance/bill/reconciliation/resource package/coupon/stored-value card/enterprise/partner). Non-Huawei Cloud or other cloud vendor billing not in scope.
Refuse Routing — Price trial / renewal quote / discount strategy / real-name authentication review results, all not in this skill scope; only one sentence pointing to console or sales-side tools, no evidence gathering, no BSS calls.
Official Identity — Does not represent Huawei Cloud; conclusions only based on evidence found at that time.
Response Language — Consistent with user; structure follows above "Response Requirements".
Environment Ready — If not ready only relay references/cli-installation-guide.md; can self-check via hcloud version / hcloud configure list, no installation or configuration on behalf of user.

References

Reference	Description
`references/semantic/catalog.yml`	Thin routing layer that maps scope/time/money_basis to billing-ontology.yml entities
`references/semantic/billing-ontology.yml`	Semantic ontology defining facts, scope, money basis, and read-only BSS operations
`references/related-commands.md`	Provides copyable BSS CLI command templates with parameter formats
`references/iam-policies.md`	Defines minimum IAM read-only permissions required for BSS queries
`references/cli-installation-guide.md`	Huawei Cloud KooCLI Installation Guide

FAQ

What does the huawei-cloud-billing-scout skill do?

How do I install huawei-cloud-billing-scout?

Run npx skills add https://github.com/huaweicloud/huaweicloud-skills --skill huawei-cloud-billing-scout in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/huawei-cloud-billing-scout/ (personal) or .claude/skills/huawei-cloud-billing-scout/ (project).

Which AI agents can use huawei-cloud-billing-scout?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is huawei-cloud-billing-scout safe to install?

Security partners have reviewed it: Gen Agent Trust Hub passed; Socket passed; Snyk passed. As with any skill, read the SKILL.md and bundled scripts before installing.

Who maintains huawei-cloud-billing-scout?

It is published by huaweicloud in huaweicloud/huaweicloud-skills. It has 17,164 recorded installs.