ClawHub

Aws Solution Architect

Security checks across malware telemetry and agentic risk

Overview

This AWS architecture skill is mostly coherent, but it includes live deployment, deletion, and cost-optimization actions without enough confirmation or safety guardrails.

Review this skill carefully before installing. It does not show malicious behavior, but only use it with explicit human approval before any AWS deploy, delete, terminate, release, or cost-optimization action. Prefer a sandbox account first, verify AWS account ID, region, stack name, change sets, backups/snapshots, and rollback plans before applying its commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill includes real deployment, modification, and deletion commands for AWS resources without an explicit warning that these actions can affect live infrastructure, destroy data, or incur charges. In an agent setting, this increases the risk that a user or downstream system runs destructive commands without confirming environment, account, region, or backup status.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The checklist recommends destructive operational actions such as deleting unused EBS volumes and stopping resources without any warning, validation criteria, or requirement for user confirmation. In an AWS architecture/cost-optimization skill, users may act on these recommendations directly, creating a realistic risk of accidental data loss, service disruption, or deletion of resources that are only intermittently used.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal