ClawHub
Back to skill

Security audit

Aws Toolkit

Security checks for vulnerabilities and agentic risk

Overview

This AWS operations skill is purpose-aligned overall, but it asks for broad cloud execution authority and includes infrastructure-changing examples without enough user approval, scoping, or rollback safeguards.

Review carefully before installing. Use this only with AWS accounts you control, least-privilege credentials, explicit account and region limits, and a dry-run or Terraform plan review before any apply, deploy, optimize, or remediation action. Avoid using automatic approval for real environments unless a human has reviewed the exact changes and rollback path.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description uses broad 'Use when' guidance such as security detection, compliance auditing, vulnerability scanning, and encryption protection without tightly defining permitted scope, required authorization, or operational boundaries. In an agent setting with exec/write tools and cloud credentials, this ambiguity can cause the system to apply powerful actions to unintended targets or contexts, increasing the risk of unauthorized or unsafe cloud operations.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The examples demonstrate infrastructure deployment and optimization actions that can create, modify, or recommend changes to AWS resources, but they do not present prominent warnings about cost, persistence, blast radius, rollback, or authorization checks immediately alongside the commands. In a skill designed for automated agent use, examples often become de facto execution patterns, so omission of strong safety gating materially increases the chance of unintended resource changes and financial impact.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
# 部署基础设施
python3 （请参考skill目录中的脚本文件） apply \
  --config ./terraform/ \
  --auto-approve
# ...
# 输出：
# VPC: 10.0.0.0/16 (2区域)
Confidence
98% confidence
Finding
This finding is a duplicate match on the same autonomous approval bypass token and represents the same underlying risk: execution of AWS changes without human confirmation. In the context of an agent skill with exec capability, that pattern is especially dangerous because a model may follow the example literally and perform real cloud modifications automatically.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
# 部署基础设施
python3 （请参考skill目录中的脚本文件） apply \
  --config ./terraform/ \
  --auto-approve
# ...
# 输出：
# VPC: 10.0.0.0/16 (2区域)
Confidence
98% confidence
Finding
This finding is a duplicate match on the same autonomous approval bypass token and represents the same underlying risk: execution of AWS changes without human confirmation. In the context of an agent skill with exec capability, that pattern is especially dangerous because a model may follow the example literally and perform real cloud modifications automatically.

Static analysis

No suspicious patterns detected.