ClawHub
Back to skill

Security audit

Verosight Monitor

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Verosight API integration, but users should be aware that searches, account identifiers, and API credentials are sent to Verosight.

Install only if you intend to use Verosight as the external monitoring provider. Avoid sending confidential investigation terms, private personal data, or production API keys unless you are authorized to share them with Verosight, and prefer environment variables or secure secret handling over command-line arguments.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documents direct shell execution via curl and local scripts, but no explicit permissions are declared to signal that command execution is required. This creates a transparency and consent problem: an agent may invoke shell-capable behavior or encourage users to do so without a clear security boundary, increasing the risk of unsafe command execution and secret handling in terminals.

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
A GET endpoint is documented as retrieving profile details while also 'auto-crawling if missing,' which introduces a hidden side effect behind a read-like operation. This is dangerous because clients, caches, or agents may treat GET as safe and idempotent, unintentionally triggering third-party crawling, data collection, rate-limit consumption, or compliance issues without explicit user intent.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README instructs users to authenticate with a live API key and send queries to an external social-media intelligence service, but it does not warn that keywords, accounts, and analysis targets will be transmitted off-platform to Verosight. In an agent-skill context, this can lead to unintentional disclosure of sensitive user data, monitoring targets, or investigation subjects, especially if an agent follows the examples automatically.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs users to authenticate and submit search terms to a third-party social monitoring service without disclosing that queries, identifiers, and retrieved monitoring data are transmitted off-platform. In this context, users may enter sensitive names, reputation issues, or investigation targets; sending that data to an external intelligence provider without warning creates privacy, confidentiality, and compliance risk.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The workflow directs collection, filtering, and reporting of social media posts, comments, account handles, engagement metrics, and 'vocal accounts' without any privacy, retention, minimization, or lawful-use guidance. In a monitoring/intelligence skill, this can normalize unnecessary profiling and redistribution of personal data, especially when compiling named accounts and quotes into reports.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The script transmits both a bearer token and user-supplied query data to a third-party API with no explicit warning, consent step, or minimization of what may be sent. In a social-monitoring context, queries can contain sensitive subjects, identities, or investigation terms, so silent outbound transmission creates privacy and operational-security risk even if this is the intended product behavior.

External Transmission

Medium
Category
Data Exfiltration
Content
fi

echo "=== SENTIMENT ANALYSIS: $QUERY ==="
curl -s "https://api.verosight.com/v1/analytics/sentiment?query=$QUERY&sources=$SOURCES&days=$DAYS" \
  -H "Authorization: Bearer $JWT" | python3 -c "
import sys, json
d = json.load(sys.stdin)['data']
Confidence
94% confidence
Finding
https://api.verosight.com/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.