ClawHub
Back to skill

Security audit

微博内容保存

Security checks across malware telemetry and agentic risk

Overview

This Weibo-saving skill is purpose-aligned but needs Review because it tells the agent to work around a network safety block and can write or sync captured content using browser and cloud credentials.

Install only if you are comfortable with an agent opening Weibo through a browser profile, saving post content and images to your Obsidian vault, and optionally sending that data to Feishu, Notion, or Jina if configured. Use a dedicated browser profile and least-privilege cloud credentials, leave optional sync tokens blank unless needed, and avoid the image-fetch workaround or broad auto-trigger behavior for private or sensitive posts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill is designed to auto-run on broad triggers like '微博链接', 't.cn', 'weibo.com', and forwarded Weibo content, which can cause unintended activation during ordinary conversation or when links are mentioned incidentally. Because the workflow performs browsing, local file writes, and optional third-party synchronization, accidental invocation can lead to unintended data collection and persistence.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill description advertises automatic fetching, local Obsidian writes, and optional synchronization to Feishu and Notion, but it does not present a clear upfront user warning about these side effects, destinations, or the sensitivity of captured content. In practice, this can cause users to trigger storage and external transmission without informed consent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The image handling rules explicitly state that files with the same name across different posts will be automatically overwritten in a shared imgs directory. This creates a real integrity risk: saved notes may silently point to the wrong image later, and previously stored data can be destroyed without notice.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.