Security audit

CDN配置工具包免费版

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a CDN guidance toolkit, but it includes live CDN account changes and host firewall rules without enough scoping or safety controls.

Install only if you want an agent to assist with CDN configuration and diagnostics, and review every command before execution. Do not let it run Cloudflare API changes or iptables rules against production systems unless you have a least-privilege token, a staging or maintenance plan, backups of current settings, and a rollback path.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (6)

Description-Behavior Mismatch

Medium

Confidence: 89% confidence
Finding: The skill is presented as a CDN configuration and optimization toolkit, but it also includes host-level firewall modification commands that directly change network access policy on the machine. This expands scope into privileged system administration and can cause unintended service disruption or be misused to alter host exposure beyond what a user expects from CDN guidance.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: The included iptables example requires elevated privileges and changes inbound access control, which is a sensitive administrative capability not strictly necessary for a general CDN toolkit. In an agent context with exec access, such commands could be applied on the host and lock out traffic or expose operators to availability incidents.

Intent-Code Divergence

High

Confidence: 86% confidence
Finding: The documentation claims the skill is not intended for operations or deployment management, yet elsewhere it provides live CDN API mutation examples and host-level network administration commands. This contradiction increases the chance that users or agents will invoke impactful operational actions under the assumption the skill is only advisory.

Vague Triggers

Medium

Confidence: 80% confidence
Finding: The trigger condition is broad enough to match generic development, debugging, and deployment requests, which can cause the skill to activate outside narrowly intended CDN scenarios. In an agent ecosystem, overbroad activation increases the risk of unnecessary exec use or application of unrelated but impactful commands.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The documentation includes authenticated Cloudflare API calls that can change zone settings but does not prominently warn about token sensitivity, account scope, or the fact that these calls modify production configuration. In an agent or copy-paste workflow, this can lead to credential exposure or unintended account-impacting changes.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The firewall rule example fetches Cloudflare IP ranges and appends iptables rules ending with a default DROP for port 443, but it lacks warnings about service lockout, incomplete rule sets, IPv6 gaps, persistence behavior, and rollback. Executing such commands on the wrong host or with missing prerequisites can immediately block legitimate HTTPS access.

Static analysis

No suspicious patterns detected.