ClawHub
Back to skill

Security audit

Alibabacloud Sas Alert Handler

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Alibaba Cloud Security Center alert-handling guide with real cloud-operation authority, so it is appropriate only for users who intentionally want an agent to help manage SAS alerts.

Install only if you want an agent to operate Alibaba Cloud Security Center alerts. Use a dedicated least-privilege RAM user or role, prefer read-only permissions until you need remediation, avoid pasting real long-lived AccessKeys into command examples or logs, and manually review any block, quarantine, kill, ignore, or whitelist action before approving it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The documentation includes realistic command examples for setting long-lived AccessKey credentials and shows they are stored in ~/.aliyun/config.json, but the warning about secret handling comes much later. In an agent or copy-paste workflow, this can normalize exposing secrets in shell history, logs, screenshots, or persisted config before readers encounter the security guidance.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The document recommends potentially destructive default actions such as blocking IPs, killing processes, quarantining files, and ignoring certain alerts, but it does not consistently warn about operational impact, false positives, or the need for verification before execution. In an alert-handling skill, these recommendations can directly influence automated or semi-automated remediation decisions, increasing the risk of service disruption, loss of forensic evidence, or unsafe suppression of real threats.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
This documentation includes high-impact alert-handling operations such as blocking IPs, killing processes, quarantining files, disabling defenses, ignoring alerts, and adding whitelist rules, but it provides no explicit confirmation, authorization, or safety guidance before performing those actions. In a security automation skill, that omission increases the chance that an agent or user will execute irreversible or security-weakening actions directly from examples, causing service disruption, loss of forensic evidence, or suppressed detections.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.