ClawHub

Clawd Presence

Security checks across malware telemetry and agentic risk

Overview

This skill is a local status display that writes its own config and status files and does not show hidden network, credential, or destructive behavior.

Install only if you are comfortable running local Python scripts that create and update config.json and state.json in the skill directory. Use manual configuration instead of --auto if you do not want it to inspect Clawd/Clawdbot config files, and keep status messages non-sensitive because they are stored locally and displayed on screen.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill documentation describes scripts that read and write configuration/status data, but the skill declares no permissions. This creates a transparency and least-privilege problem: users or orchestrators may approve or run the skill without understanding that it performs filesystem operations, which can lead to unintended access or modification of local files.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal