getskill

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:28 PM UTC
• Version:
  1.0.1

Overview

This is a disclosed skill manager that installs and updates OpenClaw skills, but users should treat the remote CLI, repositories, and installer guidance as trust-sensitive.

Install only if you trust workskills.store and the @workskills/getskill CLI. Before running install or update, review the target skill source and remember that copied skills can persist and change how OpenClaw behaves. Use custom API endpoints only when you trust the endpoint and its returned repositories.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill explicitly instructs cloning repositories and copying entire skill directories into `~/.openclaw/skills`, which modifies the local filesystem and may overwrite or replace existing content, but it does not prominently warn the user about those side effects or require confirmation. In the context of a skill installer/updater that pulls remote content, silent filesystem modification increases the risk of accidental overwrite, persistence of untrusted code/instructions, and unsafe operator assumptions.

Missing User Warnings

High

• Confidence: 96% confidence
• Finding: The skill states that on Windows it may automatically download and launch a Git installer, which is effectively remote code acquisition and execution behavior, yet the document does not require strong user consent, origin verification, or integrity checks. Because this skill also supports custom API endpoints and remote repository interactions, this makes the behavior especially dangerous: users may be induced to run downloaded installers without adequate provenance validation.

VirusTotal

63/63 vendors flagged this skill as clean.