ClawHub
Back to skill

Security audit

marketing-strategy

Security checks across malware telemetry and agentic risk

Overview

This is a coherent China-focused marketing strategy skill with a disclosed benchmark API helper and no evidence of hidden, destructive, or credential-seeking behavior.

Install this if you want China-focused marketing strategy help. Run the benchmark script only when you are comfortable contacting the listed external API, and avoid sending confidential campaign copy or business details to that service unless you trust its handling of the data.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill advertises an executable shell script (`scripts/benchmarks.sh`) and real API access, but no corresponding permissions or safety boundaries are declared. This creates a mismatch between documented capabilities and the trust model, increasing the risk that an agent could invoke shell/network behavior unexpectedly or without informed user consent.

Vague Triggers

Medium
Confidence
76% confidence
Finding
The trigger list contains broad phrases such as 'marketing strategy', 'budget allocation', and 'KPI framework' that are likely to match ordinary conversations outside the intended scope. Overbroad activation can cause unintended routing into a skill that encourages API-backed and shell-enabled workflows, expanding the chance of accidental external actions or user confusion.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The markdown explicitly describes a real external API backend and a CLI script that retrieves data, but it provides no warning, consent flow, or restriction around outbound network access. In an agent environment, this can lead to silent transmission of prompts or parameters to a third-party endpoint and normalize unsafe execution of helper scripts without adequate transparency.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.