Security audit

忘情水会问版（孟婆汤禅意情感治愈类）

Security checks for vulnerabilities and agentic risk

Overview

This is mainly a narrative-writing helper, but it explicitly describes moving blocked code/assets into Markdown to bypass upload safety controls.

Review this package before installing. The writing functionality appears purpose-aligned, but do not treat the inline JavaScript/SVG packaging pattern as a sanctioned way to evade platform file restrictions; only copy or run the embedded code after checking it in your own project context.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Ssd 2

Medium

Confidence: 98% confidence
Finding: The documentation explicitly describes converting blocked file types into inline Markdown content to '绕开上传安全拦截' and '平台上传限制', which is a direct instruction for bypassing platform security controls. Even though the embedded examples are narrative-engine code rather than obviously malicious payloads, normalizing control evasion undermines platform trust boundaries and could be reused to smuggle unsafe code or assets past scanning and review.

Static analysis

No suspicious patterns detected.