ClawHub
Back to skill

Security audit

AI智能写作助手Pro

Security checks across malware telemetry and agentic risk

Overview

The skill matches its writing-assistant purpose, but it needs Review because user-supplied topics are placed into a shell command and may be sent to Tavily without a clear privacy notice.

Review before installing. Avoid confidential topics or drafts unless you are comfortable with Tavily search receiving topic text, and do not run it with untrusted topic strings until the shell command is replaced with argument-safe execution and the file overwrite behavior is documented.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The script silently adds web-search capability by invoking another skill via execSync and passing a network API key from the environment. This expands the skill's effective privileges beyond a local writing helper, can leak user prompts/topics to an external service, and creates a trust boundary problem because execution depends on a separately installed script in a user-writable directory.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README advertises automatic hotspot/topic searching via Tavily and instructs users to configure an external API key, but it does not disclose that user-supplied topics or prompts may be transmitted to a third-party service. This creates a real privacy and data-handling risk, especially if users input proprietary, personal, or confidential writing topics under the assumption processing is local.

Missing User Warnings

Low
Confidence
78% confidence
Finding
The README shows saving generated output directly to a user-specified file path, including shell-expanded locations like ~/article.md, without warning that existing files may be overwritten or new files created. This is a legitimate but lower-severity safety issue because it can lead to accidental data loss or unintended writes on the local system.

VirusTotal

54/54 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
scripts/write.mjs:151