ClawHub
Back to skill

Security audit

skill-curator

Security checks for vulnerabilities and agentic risk

Overview

This is a read-only OpenClaw skill auditor whose local file scanning is disclosed and aligned with its purpose.

Install only if you want a local read-only audit tool for OpenClaw skills. Run it against an explicit --skills directory when scope matters, and review the built-in token allowlist because allowlisted strings will not be reported as leaks.

SkillSpector

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The script is explicitly designed to scan every skill under the workspace and read each SKILL.md file, but it does so silently and immediately. In an agent-skill context, broad file enumeration and content inspection without an explicit user-facing notice or consent can expose sensitive local content and surprise users about the scope of access.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The hardcoded default path points to a user workspace and causes the tool to operate on local data even when no path is supplied. In a skill ecosystem, defaulting to broad local inspection increases the risk of unintended data access and makes the scan scope easy to overlook.

Static analysis

No suspicious patterns detected.