ClawHub
Back to skill

Security audit

Skill Security Auditor

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed, manual command-line scanner for reviewing other OpenClaw skills, with ordinary supply-chain cautions around optional raw downloads and pattern updates.

Prefer installing through ClawHub rather than the raw curl instructions. Use this as one review aid, not proof that another skill is safe, and only run remote pattern updates if you trust the source or can verify the downloaded data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill clearly instructs users to execute a local bash script and declares shell-capable binaries (`curl`, `jq`, `grep`, `bash`), yet no explicit permission model is declared. That mismatch can mislead users and reviewers about the skill's execution capabilities and increases the risk of unsafe installation or overtrust in a security-sensitive tool.

Context-Inappropriate Capability

Medium
Confidence
83% confidence
Finding
The skill includes an update mechanism that downloads threat-intelligence data from a remote URL and writes it into a local patterns directory, but there is no integrity verification, signature checking, or trust model described. Because this is a security tool, poisoned update data could degrade detection, create false assurances, or be used to manipulate audit outcomes.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The manual installation section instructs users to fetch SKILL.md, a shell script, and a patterns file directly from remote URLs and then make the script executable, but it does not tell users to verify publisher identity, pin a specific commit, check hashes/signatures, or inspect the downloaded files first. This increases supply-chain risk because a compromised repository, typo, or MITM/proxy tampering could lead to execution of malicious code.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.