Clawhub Top

Security checks across malware telemetry and agentic risk

Overview

This skill mainly prints public ClawHub ranking reports, but its promised daily Telegram notification is under-specified and not actually implemented in the included scripts.

Install only if you want local ClawHub ranking reports and already trust the clawhub CLI in your environment. Do not enable or build the Telegram automation unless you explicitly configure the destination, understand the report payload, and are comfortable with scheduled outbound messages.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill advertises shell-based execution guidance but does not declare corresponding permissions, which weakens transparency and consent around code execution. In an agent ecosystem, undeclared execution capability can lead users or orchestrators to invoke shell actions without appropriate review, increasing the chance of unexpected local command execution.

Tp4

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding
The documented behavior does not cleanly match the described purpose, which undermines user trust and makes it harder to assess what the skill will actually do. Behavior drift such as undocumented rankings, install command output, and missing promised notification functionality can mask riskier actions or cause unsafe assumptions by users and automation systems.

Missing User Warnings

Medium
Confidence
85% confidence
Finding
The skill describes automatic Telegram notifications without a clear user-facing disclosure of what data is sent externally, when it is sent, and to which destination. Any scheduled outbound messaging creates data-sharing and privacy risk, especially if report contents or metadata could include environment-specific information beyond what the user expects.

VirusTotal

No VirusTotal findings

View on VirusTotal