Back to skill

Security audit

Huawei Cloud Infra

Security checks for vulnerabilities and agentic risk

Overview

This Huawei Cloud management skill is purpose-aligned but uses a risky local installer and cloud-admin commands that deserve review before installation.

Review this skill before installing. Only use it if you are comfortable granting a local CLI installer, persistent Huawei Cloud credential configuration, and cloud-admin command capability. Prefer manually installing KooCLI from verified Huawei Cloud instructions, use least-privilege AK/SK credentials, and require explicit confirmation before any stop, restart, delete, or other write operation.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (3)

T03 · Remote Payload Retrieval and Execution

Error
Location
scripts/install_koocli.py:23
Finding

Mutable Remote Executable Is Installed Without Integrity Verification

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/install_koocli.py:82
Finding

Archive Extraction Does Not Prevent Path Traversal or Unsafe Links

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/install_koocli.py:112
Finding

Predictable Shared Temporary Directory Enables Local Pre-Placement and Race Attacks

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The declared purpose is infrastructure query/audit/monitoring, but the skill also instructs the agent to install software locally, run installer scripts, initialize credentials, and modify the local environment. This hidden expansion of behavior is dangerous because a user asking about cloud resources could end up authorizing local code execution, downloads, and persistent configuration changes outside the stated scope.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill invokes shell commands, reads/writes configuration files, uses environment/config state, and can trigger installation, but it declares no explicit tool scope or permissions boundary. Without an allowlist, the runtime may permit broader shell, filesystem, and network actions than users expect, increasing the chance of unintended command execution or local system modification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger conditions are broad enough to match many ordinary Huawei Cloud or security-audit questions, which can cause the skill to activate unexpectedly. In this skill, accidental activation is more dangerous because activation can lead to shell execution, local installation steps, and credential/configuration workflows rather than simple text assistance.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file includes readily executable destructive cloud operations such as deleting ECS and RDS instances, with only a generic note that write operations require confirmation. In an agent skill context, this is risky because the command reference can normalize or facilitate high-impact actions without embedding explicit per-command warnings, safeguards, or approval requirements near the dangerous commands themselves.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This file contains natural-language instructions and status output that assume Chinese as the only user language. Under the stated policy, forcing a specific language without user opt-in is a locale/language policy violation unless the constraint is explicitly justified.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/install_koocli.py (reported line 143)May include surrounding context.

python
print("\n🔍 验证安装...")
    
    try:
        result = subprocess.run(
            ["hcloud", "version"],
            capture_output=True,
            text=True,