ClawHub
Back to skill

Security audit

Cloud Architect Design

Security checks for vulnerabilities and agentic risk

Overview

This skill mainly provides cloud architecture guidance, but it also requests command execution and discusses using real cloud credentials for querying and deployment without clear safety limits.

Review this skill before installing if your agent has shell access or cloud credentials. Use read-only, least-privilege cloud roles unless you explicitly intend deployment, and require manual review of any Terraform or CLI command before it touches a live account.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill claims to be a pure Markdown skill that does not require exec, yet the manifest exposes exec and the documentation also discusses command execution failures and optional cloud CLI use. This mismatch can mislead users and downstream policy engines into trusting the skill as low-risk while it still has the ability to run commands against real environments.

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The advertised purpose is architecture design and planning, but the documentation expands scope to real resource querying and deployment through cloud CLIs. That creates a hidden capability boundary where a user may expect analysis-only behavior while the skill can interact with live cloud accounts.

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
Deployment capability is not necessary for the stated function of a cloud architecture design skill, so granting or encouraging it violates least-privilege expectations. Unnecessary execution or deployment pathways increase the chance of accidental changes, cloud spend, or misuse of attached credentials.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill references optional cloud CLI credentials and command execution without clearly warning that real credentials could expose production metadata or allow changes and deployments. In a cloud architecture context, users may supply privileged credentials, making ambiguous operational guidance especially risky.

Static analysis

No suspicious patterns detected.