ClawHub
Back to skill

Security audit

deep-content-writer

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed writing-assistance skill with research and platform-style guidance, and I found no hidden code, credential use, persistence beyond installation, or destructive behavior.

Install this if you want an agent to produce research-backed long-form content. Because it may search the web as part of drafting, avoid using it with private or sensitive topics unless you are comfortable with live research, and specify the target platform, audience, and locale to prevent unwanted style conventions.

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 (4)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill directs the agent to perform web searches and external research as a mandatory step, even though the skill is presented as a generic writing tool rather than a research-enabled workflow with explicit user consent. This can cause unnecessary external data access, increase privacy and prompt-injection exposure from untrusted web content, and expand the agent's behavior beyond what the user likely expected.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The README says the skill activates automatically for broad, common requests like writing an article, blog post, newsletter, or social media thread. In agent environments with automatic skill routing, this can cause unintended invocation on many ordinary prompts, increasing the chance that the skill influences outputs when the user did not explicitly request it. The issue is a scope/control problem rather than direct code execution, but broad activation can still create unsafe or surprising behavior.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The trigger conditions are broad enough to activate on many ordinary writing requests, including loosely related tasks like turning notes into a publishable piece. Over-broad routing can cause the agent to invoke a heavyweight skill in contexts where its mandatory research, platform shaping, and stylistic constraints are inappropriate, leading to scope creep and unintended actions.

Natural-Language Policy Violations

Medium
Confidence
77% confidence
Finding
The platform guidance hardcodes Chinese-audience conventions such as idioms, internet culture references, and a specific follow prompt without requiring user opt-in. This can misalign outputs with user intent, audience, or locale, and may inject culturally specific framing into content where it is inappropriate or misleading.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.