ClawHub
Back to skill

Security audit

Alibaba Cloud Media MPS

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward Alibaba Cloud media-processing management skill, but it should be used carefully because it can operate on cloud resources and store operational evidence locally.

Install only if you intend to manage Alibaba Cloud MPS resources. Use scoped or short-lived STS credentials, verify the exact region, buckets, media IDs, and pipeline/template IDs before any write action, require explicit approval for submit, cancel, update, delete, or bucket-binding operations, and keep output/aliyun-mps-manage private or redacted.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (4)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill instructs use of AccessKey/STS credentials from environment variables or shared credential files without warning about their sensitivity, masking, or prohibitions on writing them to logs and evidence. In a skill that also writes output files and performs troubleshooting, this increases the risk of accidental credential disclosure through logs, transcripts, or persisted artifacts.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill tells the operator to retain region, pipeline/template/workflow IDs, media IDs, and request parameters in evidence files without warning that these fields may contain sensitive operational metadata or user data. Persisting detailed request parameters can expose bucket names, internal resource identifiers, media references, and potentially tokens or business-sensitive workflow details to anyone with access to the output directory.

Missing User Warnings

Low
Confidence
91% confidence
Finding
This template explicitly instructs the agent to submit and optionally cancel media processing jobs, which are state-changing remote operations, but it provides no warning, confirmation requirement, or safety boundary. In a cloud-management skill, that omission can lead to unintended job execution, cost-incurring processing, cancellation of active workloads, or disruption of media pipelines if the agent follows the template without surfacing risk to the user.

Missing User Warnings

Low
Confidence
89% confidence
Finding
The snapshot template tells the agent to submit a snapshot extraction job, which triggers a real processing action against remote media resources, but it does not warn that this is a mutating, billable, operational step. In the context of an Alibaba Cloud MPS management skill, that increases the chance of accidental execution, unexpected charges, or processing against the wrong media asset.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.