ClawHub
Back to skill

Security audit

Healthprobe

Security checks for vulnerabilities and agentic risk

Overview

HealthProbe is a small, disclosed URL health-check service; its main risk is that it can be used to contact any URL it is given.

Install only if you want a local service that makes HTTP requests to URLs supplied by the caller. Keep it bound to localhost, do not expose it publicly, avoid probing untrusted URLs, and add allowlisting or private-network/metadata-IP blocking before using it in shared, cloud, or internet-accessible environments.

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding
The skill clearly performs outbound network activity by probing arbitrary user-supplied URLs, but the manifest does not declare corresponding network permissions. This creates a capability/permission mismatch that can hide SSRF-like behavior from reviewers and platform policy enforcement, especially because the skill is explicitly designed to reach arbitrary endpoints.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The endpoint performs outbound requests to arbitrary user-supplied URLs with redirects enabled and no visible restriction or warning, which creates an SSRF-style network probing primitive. In an agent context, this can be abused to scan internal services, access cloud metadata endpoints, or make unintended outbound requests from a trusted network location.

Static analysis

No suspicious patterns detected.