ClawHub

Viral Copy Generator

Security checks across malware telemetry and agentic risk

Overview

This is a purpose-aligned browser copywriting tool, but users should know that translation and voice features send text to third-party services despite some local-only wording in the docs.

Install only if you are comfortable with generated or entered text being sent to MyMemory for translation and tts.wangwangit.com for voice synthesis. Treat the local-only claims as applying to saved history, not to translation or TTS processing.

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The documentation asserts that translation requests use POST to avoid exposing user text in URLs, but the only endpoint shown is MyMemory's /get API and no implementation is provided to prove the request is actually sent safely. This creates a privacy and trust risk because users may submit sensitive draft content believing it will remain out of URLs, logs, browser history, and intermediary systems when that may not be true.

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The guide claims that all data is stored locally and that no user information is uploaded, but elsewhere it states that TTS and translation rely on remote services. That means user-entered text, and possibly generated content, may be transmitted off-device, creating a misleading privacy representation that can cause users to disclose sensitive data under false assumptions.

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The UI/privacy text claims the tool is a pure frontend app where all data stays only in the browser, but the code explicitly sends user-entered content to third-party translation and TTS services. This is a real security/privacy issue because users may rely on the claim and submit sensitive product, business, or personal data under a false expectation of local-only processing.

Intent-Code Divergence

Low
Confidence
85% confidence
Finding
The comment asserts that using POST avoids exposing user input in the URL, but the code still uses a '/get' translation endpoint and provides no evidence that upstream services, logs, or intermediaries will treat the request as private. While this is less severe than direct exfiltration, it is still misleading security documentation that can cause developers and users to overestimate privacy protections.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The README makes a strong privacy claim that data is only stored locally and not uploaded to servers, yet it also documents use of external TTS and translation APIs. That mismatch can mislead users into entering sensitive text under a false expectation of on-device processing, causing unintended disclosure of their content to third-party services.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are very broad and map to common, everyday requests like '生成文案' and 'TTS', which increases the chance the skill is invoked when the user did not explicitly intend to use it. Unintended invocation can cause confusing behavior and, in this skill's case, may also route user content into downstream features such as translation or TTS unexpectedly.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly references use of external Edge TTS and translation services but does not warn users that their text may be transmitted to third parties. This creates a privacy and data-handling risk, especially if users submit product plans, drafts, marketing strategy, or other sensitive business content assuming it remains local.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The manifest explicitly declares network use for TTS and translation, but it does not clearly warn users that text they enter may be transmitted to third-party services. In a content-generation skill, users may paste unpublished marketing copy, scripts, or sensitive business data, so silent external transmission creates a real privacy and data-handling risk.

External Transmission

Medium
Category
Data Exfiltration
Content
}
    },
    "translate": {
      "apiUrl": "https://api.mymemory.translated.net/get"
    },
    "storage": {
      "key": "copyRecords",
Confidence
88% confidence
Finding
https://api.mymemory.translated.net/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal