ClawHub

Clawhub Release Auditor

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed ClawHub release-auditing helper that reads a user-chosen skill folder and runs local validation or inspect commands, with publishing gated on user confirmation.

Install only if you are comfortable with the skill reading the skill directory you point it at and running local Python, OpenClaw, and ClawHub CLI commands. Before approving any publish step, confirm the active ClawHub account, target skill path, slug, and intended version.

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
92% confidence
Finding
The skill instructs the agent to read local files and execute shell commands (`python3 ...`, `clawhub inspect`) but does not declare corresponding permissions/capabilities. That creates a transparency and policy-enforcement gap: the skill appears less privileged than it really is, which can bypass user expectations, break sandboxing assumptions, or cause risky execution in environments that rely on declared permissions for review.

VirusTotal

55/55 vendors flagged this skill as clean.

View on VirusTotal