ClawHub
Back to skill

Security audit

Content System Wechat Formatter

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a formatter, but its runtime imports unreviewed Python code from a hard-coded local user path outside the package.

Install only if you trust and can inspect the local wechat_html_renderer code at the hard-coded path, or prefer a version that vendors that helper or declares a pinned, reviewable dependency. Treat the writing-pack sidecar as article input and use it only from the same trusted draft directory.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Low
Confidence
84% confidence
Finding
The manifest says it may 'optionally consume a sibling writing-pack sidecar' but does not define how that sidecar is discovered, authenticated, or constrained. In an agent environment, ambiguous sidecar resolution can cause unintended nearby files to be ingested, creating a prompt/input injection surface or accidental data exposure if attacker-controlled artifacts are present.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.