ClawHub
Back to skill

Security audit

AIMA Doctor

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a real OpenClaw repair skill, but it automatically installs and runs an external helper with broad local repair authority.

Install only if you trust the publisher and the referenced release. Before running it, verify the checksum, review the downloaded helper scripts if practical, and expect it may modify OpenClaw configuration or device/token recovery state. Use a constrained or disposable environment for sensitive installations.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill is user-invocable and described broadly as a rescue tool for when the agent is 'stuck, confused, or failing,' which can match many ordinary troubleshooting situations. That broad trigger surface increases the chance the skill is invoked inappropriately and leads to installation/execution of a helper runtime with local repair capabilities.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to automatically download, verify, and extract a runtime from GitHub into a persistent local tools directory, but it does not require an explicit user-facing warning or consent at the moment of modification. Even with checksum verification, this is still remote code acquisition and local installation, which materially changes the host and may surprise users or violate least-astonishment and approval expectations.

Static analysis

No suspicious patterns detected.