Back to skill

Security audit

Huawei Cloud Swr Image Automation

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Huawei Cloud SWR automation guide, but users should treat its sync and auto-deploy commands as real cloud changes.

Install only if you intend to let the agent help administer Huawei Cloud SWR. Use least-privilege IAM credentials, prefer test namespaces and non-production clusters first, avoid broad tag patterns like .* for production, and require an explicit review before any create, update, delete, syncAuto=true, override=true, or trigger enablement command.

SkillSpector

By NVIDIA
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 (7)

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The skill provides a destructive delete command for image sync configuration without an explicit confirmation or warning step. In an agent context, this increases the risk of irreversible operational changes being carried out from ambiguous or mistaken user requests, especially because auto-sync is persistent and affects production image distribution.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The trigger deletion command removes deployment automation and can disrupt release workflows or accidentally stop/update production behavior if invoked incorrectly. Because triggers directly affect downstream CCE/CCI deployments, omission of an explicit warning and confirmation makes accidental destructive action more dangerous in this skill than in a read-only administrative tool.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The document promotes creation and management of SWR triggers that automatically update CCE/CCI workloads when images are pushed, but it does not prominently warn that this enables production-affecting automatic deployments. In a skill intended for infrastructure automation, omitting an explicit safety warning and approval guidance can lead users to enable high-impact behavior without understanding rollout, availability, or supply-chain risks.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The verification guide includes create and delete operations against real SWR repositories as routine validation steps, but it does not explicitly warn that these commands modify cloud state and may incur operational or cost consequences. In a skill intended for automation and cloud operations, users may execute the steps verbatim in production-like environments, causing unintended replication configuration changes or deletion of existing sync settings.

Missing User Warnings

High
Confidence
97% confidence
Finding
The trigger management section documents trigger creation, update, and deletion without warning that these actions can change deployment behavior in connected CCE environments and may automatically roll out workloads on image events. Because triggers bridge repository events to cluster actions, a user following the guide could unintentionally enable or disrupt live deployments, making the operational blast radius significantly higher than ordinary repository configuration changes.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
### W1: Create a Trigger

Set up auto-deploy to a CCE workload when new images are pushed:

**Pre-creation Checklist**:
1. Verify repository exists:
Confidence
85% confidence
Finding
auto-deploy

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Common Scenarios

### S1: CI/CD Auto-deploy Pipeline

Set up automatic deployment updates when CI pushes new images:
Confidence
87% confidence
Finding
Auto-deploy

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.