ClawHub
Back to skill

Security audit

Creator Alpha Feed

Security checks across malware telemetry and agentic risk

Overview

This skill openly describes a workflow that collects AI-related X posts, sends a short digest to a configured group channel, and writes a full Markdown report to Obsidian.

Before installing, confirm the Obsidian Vault path, whitelist accounts, keywords, and group webhook/chat destination. Run it only when you are comfortable sharing the selected X content summary with that channel, especially if your X session can view non-public or account-specific content.

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
83% confidence
Finding
The example trigger phrase “采集今日 AI 内容日报” is broad and lacks safeguards such as confirmation, scope preview, or explicit mention of side effects. Because this skill can scrape logged-in content, write files, and push messages externally, an underspecified trigger increases the chance of accidental activation of a workflow with meaningful side effects.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly writes reports into an Obsidian Vault but does not present a prominent warning that local files and directories will be created or modified. This is dangerous because users may invoke the skill without realizing it performs persistent local changes, which can overwrite data, create sensitive artifacts, or write into an unintended Vault path if configuration is incorrect.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill transmits collected content to a group channel via webhook or chat integration without a clear privacy warning or consent boundary. This is risky because scraped content, account metadata, links, and possibly user-specific or logged-in-view content could be forwarded to external recipients unintentionally, creating confidentiality, compliance, and data-handling issues.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.