Security audit

智享写作助手·物哀体系

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese writing humanizer with no executable code, but it explicitly encourages fake or semi-fictional provenance cues to make writing appear human.

Install only if you specifically want this style of Chinese creative-writing rewrite and can review every change. Do not use its provenance, semi-fictional detail, or anti-detection guidance for factual, academic, professional, legal, or public-facing writing unless all sources and authorship claims are real and disclosed.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

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
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Context-Inappropriate Capability

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill explicitly encourages adding provenance-like artifacts such as handwritten scans, meeting transcripts, unpublished preprints, and semi-fictional insider information to make text appear more human. That crosses from stylistic editing into manufacturing credibility signals and potentially deceptive evidence, which can facilitate fraud, misinformation, or evasion of authenticity checks.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The description and trigger metadata explicitly position the skill for Chinese writing only, using phrases like “中文创意写作” and Chinese-only triggers, but the document does not state this as an opt-in language choice or explain a justified regional restriction. This can violate language/locale policy when a skill implicitly forces a specific language without user selection.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The markdown declares Write and Edit as allowed tools, and the workflow/output sections indicate the skill will rewrite user text, but there is no user-facing warning that the skill may alter provided content. For markdown files, behaviors affecting user data should be disclosed so users understand that input text may be transformed or overwritten.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The manifest trigger list includes generic phrases such as “润色”, “中文写作”, and “故事改写”, which are common everyday writing requests rather than narrowly scoped activation phrases. The file does not provide exclusion conditions or negative examples to clarify when the skill should not activate, increasing the risk of unintended invocation.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The skill claims to improve and humanize writing, but it also recommends deliberate errors, personal 'error fingerprints,' and semi-fictional details. This contradiction makes the tool dangerous because it normalizes deceptive modifications that can mislead readers about authorship, provenance, or factual grounding.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.