Security audit

深度解析Skill技能，输出四部分报告：功能摘要、场景分析、流程图、评分点评，让你一眼看懂优秀技能背后的设计逻辑ub 链接。

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only skill that analyzes user-provided skill packages, with no hidden execution or persistence found.

Install if you want an agent to summarize and critique skill packages. Provide only the specific package, link, or path you want inspected, and avoid archives or folders containing secrets or unrelated private files.

SkillSpector

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The trigger phrases are broad, conversational, and overlap with normal discussion about skills or packages, so the skill can activate when the user is merely asking a question rather than intentionally invoking tooling behavior. Unintended invocation can cause the agent to process links, paths, or uploaded files unexpectedly, increasing the chance of unwanted data access or confusing behavior.

Vague Triggers

Medium

Confidence: 96% confidence
Finding: The activation logic says to start when the user says any listed phrase, but it does not sufficiently constrain context, user intent, or allowed input sources before proceeding into file/link handling. In an agent environment, ambiguous auto-start behavior can lead to accidental processing of local paths or remote content that the user did not clearly intend to submit for analysis.

Static analysis

No suspicious patterns detected.