Setup Multi Gateway

• Outcome: Review
• Latest audit: May 29, 2026, 9:26 PM
• Version:
  1.2.0

Overview

The skill mostly matches its gateway-setup purpose, but it embeds a shared API key and writes secrets into generated local configuration without clear disclosure.

Review the wizard before installing or running it. Do not use it with production OpenClaw or Feishu configurations unless you are comfortable with plaintext local secrets, copied skills/memory, public IP-discovery requests, and persistent user-level services. The embedded API key should be removed and treated as exposed before trusted use.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Intent-Code Divergence

High

• Confidence: 99% confidence
• Finding: The script hardcodes an API key and later injects it into generated models.json files, which creates a credential exposure and unauthorized-use risk. Any user with access to the script, generated files, backups, logs, or source control can recover the key and use the associated external service at the owner's expense or to access protected resources.

Intent-Code Divergence

Medium

• Confidence: 97% confidence
• Finding: Replacing $api-key with a built-in secret during file copy silently alters model credentials and propagates the same secret to every generated agent configuration. This increases blast radius, makes secret rotation difficult, and can expose a privileged shared credential across multiple workspaces and users.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: A hardcoded API key embedded in the script is a direct secret-management failure, and the wizard copies it into generated configuration without prominently warning the operator. In this admin-style setup tool, that behavior is especially dangerous because it encourages broad credential reuse and leaves the key in multiple plaintext locations.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The IP discovery routine contacts third-party services such as ifconfig.me, api.ipify.org, and icanhazip.com without explicit consent or notice. While the data sent is limited, it leaks host metadata and creates an unexpected outbound network dependency in a local configuration wizard.

VirusTotal

61/61 vendors flagged this skill as clean.