Security audit

深入浅出讲明白

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese writing-template skill for explaining complex topics, with no executable code or hidden install behavior.

Install this only if you want a reusable Chinese article-writing framework. Because it prompts the agent to research topics and may include links in generated articles, review sources and outputs before relying on or publishing them; the Ralph Wiggum example should not be treated as safety guidance for running autonomous coding tools.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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 (1)

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The content explicitly promotes unattended autonomous code iteration ('全程不需要人工干预') and encourages users to walk away while the agent continues modifying and validating code. In a coding-agent skill, this can normalize unsafe operation without mentioning safeguards such as sandboxing, branch isolation, test/environment boundaries, or protection of production data and secrets.

Static analysis

No suspicious patterns detected.