huawei-cloud-cdn-abnormal-status-code-analysis

• Outcome: Pass
• Latest audit: Sep 3, 2026 · 12:55 PM UTC
• Version: 1.0.0

Overview

This is a coherent read-only Huawei Cloud CDN troubleshooting skill, with ordinary caution needed for cloud credentials and access logs.

Install only if you administer the relevant Huawei Cloud CDN tenant. Use a least-privilege read-only hcloud profile, do not paste AK/SK into chat, keep downloaded CDN logs local and access-controlled, and redact or delete log output when it is no longer needed.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (12)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill clearly instructs use of shell commands and outbound network access (hcloud CLI, Python requests log fetch), but it does not declare any explicit tool scope such as allowed-tools or permissions. That creates a governance gap: an agent runtime may grant broader capabilities than reviewers or users expect, increasing the risk of unintended command execution or external data access.

Vague Triggers

Medium

• Confidence: 85% confidence
• Finding: The trigger list includes very broad phrases like '4xx', '5xx', 'status code', and 'abnormal status', which can match many generic troubleshooting conversations outside the intended Huawei Cloud CDN context. Overbroad activation can cause the skill to run shell/network diagnostics in situations where the user did not intend this specialized workflow, increasing the chance of unnecessary data access or confusing unsafe tool use.

Autonomous Decision Making

Medium

• Category: Excessive Agency

Content

> **Pre-check: Python `requests >= 2.25`** — `python -c "import requests; print('ok')"`. The log helper downloads via `requests`.
> Install if missing: `pip install requests>=2.25`

> **Pre-check: hcloud credentials** — `hcloud configure list`. If invalid, stop and guide secure configuration. Never read/echo/print AK/SK; never ask the user to paste AK/SK in chat.

> **⚠️ hcloud parameter format**: all params must use `--key=value`.

• Confidence: 80% confidence
• Finding: Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The documentation explicitly instructs operators to retrieve and process CDN access logs containing client IP addresses, user agents, requested URLs, and timestamps, but provides no privacy, retention, minimization, or access-control guidance. In an incident-response skill this data use can be legitimate, yet without safeguards it increases the risk of unnecessary collection, overexposure, or mishandling of personal and potentially sensitive request data.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The verification flow retrieves CDN log listings and downloads log content for forensic analysis without warning that access logs may contain sensitive operational data, client IPs, URLs, headers, or other user-related telemetry. This creates a real risk of unnecessary exposure of production traffic data during routine verification, especially if logs are fetched to insecure local environments or shared in reports.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: The documentation specifies daily intervals as "CST 0:00 points" without offering a user choice or clearly framing this as a fixed service-side API constraint. This can create a language/locale policy concern because it imposes a locale-specific time interpretation in the skill instructions.

Natural-Language Policy Violations

Low

• Confidence: 86% confidence
• Finding: The document states 'use --cli-region=cn-north-1' as a fixed requirement. Under the policy, forcing a specific locale or region should either be optional, user-selected, or clearly justified as region-specific; the parenthetical note about cn-north-4 not being supported explains one exclusion but does not fully document why the skill must be constrained to this region overall.

Natural-Language Policy Violations

Low

• Confidence: 88% confidence
• Finding: The note states that all hcloud commands use `--cli-region=cn-north-1`, which imposes a fixed locale/region choice in natural language. The file does not offer user opt-in or explain why this region must be used, so it can violate the policy against forcing a specific locale without justification.

Natural-Language Policy Violations

Low

• Confidence: 90% confidence
• Finding: The document includes Chinese-only phrases such as "刷量" in its core instructions and decision criteria, and later uses Chinese wording again in the exit criteria. Because the file does not offer an alternate language or indicate that Chinese is a required locale, this may violate a language/locale policy requiring user choice or opt-in.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: The document instructs users to align timestamps to "non-CST-0:00" and later to "whole-hour CST points," which imposes a specific timezone in natural-language guidance. The file does not explain why CST is required by the service API or offer user choice, so this can conflict with language/locale policy expectations.

Natural-Language Policy Violations

Low

• Confidence: 75% confidence
• Finding: The line tells users to align queries to "whole-hour CST points," again enforcing a specific locale/timezone convention in user-facing guidance. Because no rationale is provided in the file, this reads as a hard locale constraint rather than a documented service requirement.

Missing User Warnings

Low

• Confidence: 93% confidence
• Finding: The verification procedure instructs the operator to inspect and use an existing AK/SK-backed hcloud profile, but it does not warn that the skill will operate with live cloud credentials against real tenant resources. In a security-sensitive environment, this can lead users to run diagnostic actions without understanding credential scope, account impact, or audit implications.

Static analysis

No suspicious patterns detected.