ClawHub
Back to skill

Security audit

ZM 公众号草稿发布与核验

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed WeChat draft-publishing helper that uses expected credentials and network calls to create and verify drafts, with no evidence of hidden exfiltration or destructive behavior.

Install only if you intend to let the agent create WeChat Official Account drafts. Use it with a dedicated WeChat AppID/AppSecret, review article and image paths before running, and remember that content and cover images will be sent to WeChat even though the script is scoped to draft creation rather than mass publication.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
94% confidence
Finding
The skill advertises and instructs use of capabilities that read configuration and article files, write publishing artifacts, access environment-provided credentials, and make network calls to WeChat, but it does not declare corresponding permissions. That creates a trust and policy gap: a reviewer or runtime may underestimate what the skill can access and transmit, increasing the chance of unintended secret exposure or unauthorized data handling.

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The verification block independently reads WeChat app credentials from local config or environment and performs direct API calls, which expands the skill's privilege beyond simple orchestration of the md2wechat tool. In an agent setting, this creates a broader secret-access and network-capable path than the user may expect, increasing the blast radius if the script is invoked on an untrusted system or modified workflow.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The file advertises automatic image upload and one-click publishing to a public platform without warning the user that local files and article content will be transmitted to external services. In a publishing skill, this can cause unintended disclosure of unpublished content, metadata, or local assets if a user runs the documented workflow without understanding the network and posting side effects.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal