ClawHub
Back to skill

Security audit

Hk Stock Morning Report

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a coherent Hong Kong stock report generator, but it can send an internal-labeled report to WeChat and Feishu without clear per-send user confirmation or recipient scoping.

Review before installing if the agent has messaging tools connected. Require a preview and explicit approval before any WeChat or Feishu send, confirm the exact recipient/group, and ensure the generated report contains only information approved for those channels.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill explicitly directs reading and writing local files and using network sources, but no permissions are declared. That creates a capability/permission mismatch: operators and users cannot accurately assess what the skill may access or modify, increasing the chance of unintended file overwrites or network use without proper review.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrases are common finance terms that can match ordinary user requests, increasing the risk of accidental activation. In this skill's context, accidental activation is more concerning because it may initiate network lookups, file reuse, and outbound report workflows without sufficiently clear user intent.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill describes sending generated content to WeChat personal and Feishu groups but does not require an explicit user-facing warning or confirmation before transmission. Because the report is labeled internal and may include sensitive trading-desk material, silent or ambiguous external sending could cause confidentiality breaches or unintended dissemination.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
This file hard-codes the report format and content in Chinese for a Hong Kong stock morning report, with no indication that language should adapt to the user's preferences. For a reusable agent skill, forcing a specific language/locale without user opt-in can cause misuse, user confusion, or non-compliant outputs when the user expects another language or region-specific framing.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.