Webhook Relay

• Outcome: Pass
• Latest audit: Jun 15, 2026 at 7:01 PM UTC
• Version:
  1.0.0

Overview

This skill is a coherent Webhook Relay guide, but users should treat its public tunneling and webhook capture workflows as sensitive network exposure.

Install only if you intend to use Webhook Relay. Before running tunnel or forward commands, confirm the destination is safe to expose, use authentication where appropriate, avoid exposing admin panels, databases, SSH, or internal dashboards unless you fully understand the risk, and remove buckets, tunnels, functions, crons, or background services when finished.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The skill description is broad enough to match many generic webhook, tunneling, forwarding, and debugging requests, increasing the chance it is invoked in contexts where the user did not explicitly ask to expose internal services or transmit traffic externally. In an agent setting, over-broad triggering can cause risky networking guidance to appear unexpectedly and normalize exposure of private endpoints.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill prominently explains how to expose localhost, LAN, Kubernetes, and other internal services to the public internet, but does not put an equally prominent warning near the top about unintended public access, authentication, data exposure, and scope restriction. Because the core purpose is crossing trust boundaries, weak safety framing increases the risk that users publish sensitive internal services without understanding the consequences.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The example includes inline basic-auth credentials directly on the command line, which can leak via shell history, process listings, terminal scrollback, or copied documentation. While the example is for a demo, this pattern encourages insecure secret handling that can lead to credential disclosure.

Missing User Warnings

Low

• Confidence: 80% confidence
• Finding: The bin is described as public and temporary, with warnings about secrets and PII, but it does not clearly warn that arbitrary third parties can send traffic to the endpoint and pollute captures, trigger mock responses, or interfere with testing. This can mislead users into trusting captured requests as originating only from the intended provider.

VirusTotal

63/63 vendors flagged this skill as clean.