ClawHub
Back to skill

Security audit

huawei-cloud-lts-log-inspector

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documented Huawei Cloud log inspection and export helper that needs cloud log access and requires confirmation for export creation or deletion.

Use this skill only with the intended Huawei Cloud account and region. Prefer the documented read-only IAM policy unless you need OBS export, confirm timezone and region before running reports, and review cost/storage impact before approving any transfer creation or deletion.

A.I.G

Vulnerability Patterns
  • 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
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill documents extensive shell-based hcloud CLI operations but does not declare any explicit tool scope, permissions, or allowed-tools boundary. That creates an unsafe trust gap: an agent may execute shell commands for read access, export creation, and deletion without a machine-readable restriction layer, increasing the risk of over-broad command execution or misuse of authenticated cloud credentials.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The command examples hard-code --timezone=Asia/Shanghai, which imposes a specific locale setting in the workflow documentation. The file does not indicate that this timezone is optional, user-selectable, or justified as a region-specific constraint, so it may violate language/locale policy expectations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

This diagnostic workflow again specifies --timezone=Asia/Shanghai as a fixed value, reinforcing a locale-specific default without offering user choice. Because the skill is otherwise presented as generally applicable, the fixed timezone appears to be an undocumented locale constraint.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The full patrol report example includes --timezone=Asia/Shanghai without any opt-in or explanation for why that locale is mandatory. This creates an avoidable locale policy issue in natural-language guidance.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The file explicitly authorizes autonomous execution of LTS query commands 'without confirmation', which reduces user oversight before interacting with a cloud environment. Although the commands are read-only, they can still enumerate sensitive infrastructure metadata and log information, so the danger is unauthorized disclosure or overly broad data access if the agent is triggered in the wrong context or with excessive privileges.

Content

Scanner excerpt · references/verification-method.md (reported line 15)May include surrounding context.

Query Commands (Read-Only)

Query commands are safe to execute without confirmation. Verify results by checking the response structure:

bash

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This manifest-like JSON sets the default region to "cn-north-4" and later fixes the timezone to "Asia/Shanghai". Forcing a specific region/locale in natural-language/config values without user opt-in can violate language/locale policy expectations, especially when no alternative or justification is provided.

Content

No source excerpt is available for this finding.