Security audit

品牌标识设计免费版

Security checks for vulnerabilities and agentic risk

Overview

This brand-design skill is mostly a Markdown workflow, but it asks for command execution, file writing, callback URLs, and broad activation without clear safety limits.

Review before installing. Use this only in a constrained workspace and avoid providing callback URLs unless you trust the destination. Prefer a version that removes shell execution, narrows activation to brand identity tasks, and limits any file writes to explicit user-requested exports.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Context-Inappropriate Capability

Medium

Confidence: 91% confidence
Finding: The skill advertises exec and write capabilities even though its stated purpose is brand identity design, which normally only needs content generation and perhaps read access. This unnecessary capability expansion increases the attack surface and could let prompt-influenced workflows create or modify files or run local commands without a clear business need.

Context-Inappropriate Capability

Medium

Confidence: 83% confidence
Finding: Adding callback_url support introduces outbound communication capability that is not justified by a logo/brand design skill. In an agent context, this can be abused for data exfiltration, SSRF-style requests, or sending generated content and local context to attacker-controlled endpoints.

Context-Inappropriate Capability

Medium

Confidence: 86% confidence
Finding: The documentation instructs use of ping/network diagnostics unrelated to the core branding function, signaling that the skill may induce network-active behavior outside scope. Such behavior can be repurposed to probe connectivity, interact with internal addresses, or normalize unnecessary command execution in a low-trust content workflow.

Vague Triggers

Medium

Confidence: 78% confidence
Finding: The activation scope is overly broad and malformed, which can cause the skill to trigger in contexts far beyond brand design. Overbroad routing increases the chance that higher-risk tool permissions like exec/write are invoked in unrelated conversations, amplifying misuse risk.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill exposes system-affecting capabilities such as exec and write without clearly warning users or defining safety boundaries. In practice, this can mislead operators into invoking a design skill that performs local actions, increasing risk of unintended file changes or command execution from untrusted prompt content.

Static analysis

No suspicious patterns detected.