ClawHub

Skill Publish

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed ClawHub publishing helper that checks a user-selected skill folder and only publishes after user confirmation.

Install this only if you intend to prepare or publish OpenClaw skills to ClawHub. Before approving any publish, verify the target folder, slug, display name, version, changelog, tags, and active `clawhub whoami` account; avoid giving the agent tokens unless you explicitly want it to handle local CLI authentication.

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
93% confidence
Finding
The skill instructs the agent to read arbitrary local skill folders and execute shell commands such as `python3 .../check_publish_ready.py`, `clawhub publish`, and `clawhub search`, but it does not declare permissions for file reads or shell execution. That mismatch is a real security issue because users and platform controls may not have clear visibility that the skill can inspect local files and invoke publishing commands that interact with external services.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal