ClawHub

cloud-api-resource

Security checks across malware telemetry and agentic risk

Overview

This skill is openly for creating UCloud cloud resources, but its broad trigger instructions could cause costly cloud actions from ambiguous requests.

Install only if you want an agent to help create UCloud resources. Use limited-permission subaccount keys, require an explicit UCloud/provider confirmation before any API call, review the full resource and cost summary carefully, and avoid giving long-lived or broad private keys in chat.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger logic is overly broad: the workflow says the agent should act whenever a user expresses a generic resource-creation intent, then infer the resource type and proceed. In an agent environment with multiple skills, this can cause unintended invocation of a high-impact cloud provisioning skill from ambiguous natural-language requests, leading to unauthorized or accidental resource creation and billing.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The interactive-mode trigger examples include generic phrases like '创建云主机' and '帮我创建一台服务器' without any UCloud qualifier. Because this skill performs real external side effects, such broad phrasing increases the chance that ordinary conversation or requests intended for another provider will route into this skill and create cloud resources unintentionally.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The listed trigger phrases contain generalized language like '创建 UCloud 存储桶' and especially '调用 UCloud API 创建资源', combined with the statement that any intent to create UCloud resources can invoke the skill. For a skill capable of provisioning infrastructure and API keys, permissive triggering materially raises the risk of accidental invocation, privilege misuse, and unexpected external actions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal