ClawHub
Back to skill

Security audit

Social Media Monitor

Security checks across malware telemetry and agentic risk

Overview

This skill is a local CSV-based social media monitoring tool, and its local file writes match its documented reporting and settings features.

Install only if you are comfortable running a local Node MCP server that stores monitoring keywords, alert settings, CSV-derived results, and generated Markdown reports in the skill directory. Avoid putting highly sensitive business data in the CSV files unless local persistence is acceptable, and review/pin npm dependencies if your environment requires strict reproducible builds.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (7)

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The skill goes beyond passive analysis by persistently writing reports, keyword lists, and configuration files to disk. In an MCP/tooling context, unexpected state changes are security-relevant because they can alter future behavior, leave sensitive artifacts on the host, and violate least-privilege expectations for a monitoring/analysis skill.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The weekly report tool writes a Markdown file to local storage without an explicit user-facing warning that invoking the tool will modify the filesystem. In agent environments, silent writes are dangerous because users may expect analysis-only behavior, and generated reports can persist sensitive operational data on disk.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The keyword and alert-threshold tools silently persist monitoring configuration to JSON files, changing future tool behavior across sessions. Hidden persistence is security-relevant because it creates durable state that users may not realize exists, and that state can influence later analyses or leave behind sensitive business-monitoring terms.

Unpinned Dependencies

Low
Category
Supply Chain
Content
"author": "小爪 🦞",
  "license": "MIT",
  "dependencies": {
    "@modelcontextprotocol/sdk": "^1.27.1",
    "keyword-extractor": "^0.0.28",
    "sentiment": "^5.0.2",
    "zod": "^4.3.6"
Confidence
88% confidence
Finding
"@modelcontextprotocol/sdk": "^1.27.1"

Unpinned Dependencies

Low
Category
Supply Chain
Content
"license": "MIT",
  "dependencies": {
    "@modelcontextprotocol/sdk": "^1.27.1",
    "keyword-extractor": "^0.0.28",
    "sentiment": "^5.0.2",
    "zod": "^4.3.6"
  }
Confidence
84% confidence
Finding
"keyword-extractor": "^0.0.28"

Unpinned Dependencies

Low
Category
Supply Chain
Content
"dependencies": {
    "@modelcontextprotocol/sdk": "^1.27.1",
    "keyword-extractor": "^0.0.28",
    "sentiment": "^5.0.2",
    "zod": "^4.3.6"
  }
}
Confidence
84% confidence
Finding
"sentiment": "^5.0.2"

Unpinned Dependencies

Low
Category
Supply Chain
Content
"@modelcontextprotocol/sdk": "^1.27.1",
    "keyword-extractor": "^0.0.28",
    "sentiment": "^5.0.2",
    "zod": "^4.3.6"
  }
}
Confidence
87% confidence
Finding
"zod": "^4.3.6"

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.