Back to skill

Security audit

fanqie-masterclass

Security checks for vulnerabilities and agentic risk

Overview

The skill is a non-executable Chinese web-fiction course, but it promotes questionable copying and fake engagement tactics and includes sensitive tax-ID guidance without safeguards.

Review this skill before installing if you want only general writing help. It is best treated as a Chinese Fanqie web-fiction training resource, not a universal writing assistant. Avoid using its imitation and fake-engagement advice, and do not enter real ID, address, banking, or tax information into chat or unofficial tools; use only official platform forms and qualified tax guidance for those steps.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (17)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description says it should be used whenever a user wants to 'learn writing,' 'improve short-form writing,' 'diagnose work problems,' 'generate titles/outlines,' or 'polish AI tone,' which overlaps heavily with ordinary writing-assistant requests. In an agent environment, such broad activation criteria can cause unintended routing to this skill for generic content-help tasks, leading the skill to steer outputs toward a specific commercial writing formula and AI workflow when the user did not explicitly ask for this domain-specific behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The quick-start routing maps very generic phrases like '想学习/怎么写,' '帮我生成,' '诊断,' and '润色' directly into skill modes. Because these phrases are common across everyday writing support, the agent may activate this skill for unrelated requests, creating prompt-routing confusion and causing domain-inappropriate advice or content transformation without clear user consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This markdown file is entirely prescriptive about writing style in Chinese, using phrases like '必须删除或替换' and '必须替换' without any indication that the guidance is optional or language-specific. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The entire skill content is written only in Chinese, with no indication that users may choose another language or that the skill is intended exclusively for a Chinese-speaking or region-specific audience. Under the policy, language constraints should be opt-in or clearly justified rather than implicitly imposed.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The entire skill guidance is written as prescriptive content in Chinese and does not indicate that users may choose another language or locale. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The entire skill content is written as prescriptive guidance only in Chinese, with no indication that users may choose another language or locale. Under the policy rule, forcing a specific language without opt-in is a natural-language policy concern unless the locale restriction is clearly documented and justified.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This section explicitly teaches how to imitate successful works while evading plagiarism detection by preserving the 'core logic' and changing only surface elements. That meaningfully increases the risk of copyright infringement, deceptive content reuse, and downstream misuse of the skill for policy evasion rather than legitimate writing education.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The document recommends manipulative engagement tactics such as using sockpuppet-style comments ('小号发') to manufacture interest and steer user behavior. This promotes deceptive platform-gaming behavior that can violate platform integrity rules and normalize inauthentic amplification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The content instructs users to enter highly sensitive personal and tax data, including a real name, residential address, and Chinese national ID number, but provides no privacy warning, secure-handling guidance, or caveat that platform tax forms should only be completed on trusted official sites. In a skill context, users may copy this information into external tools or untrusted intermediaries, increasing risk of identity theft, data leakage, or phishing-assisted fraud.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file presents all instructions and checklist items exclusively in Chinese, with no indication that users may choose another language or that the Chinese-only constraint is required for a documented regional purpose. That can violate a language/locale policy requiring user opt-in or explicit justification for a forced language.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file is instructional content for fiction writing, but it explicitly encourages replacing emotion words with violent actions such as throwing objects, choking, stabbing, and humiliating others. In a general-purpose writing skill without a content warning or safety boundary, this can normalize or amplify violent imagery and may cause distress or be repurposed to generate harmful content on demand.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The line includes a self-harm example ('想从楼上跳下去') as a way to depict embarrassment, without any warning or contextual safeguard. Even in creative-writing guidance, casually presenting self-harm as an expressive device can be triggering and may lead the skill to surface unsafe language in normal user interactions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The entire document is written as prescriptive guidance in Chinese and presents no option for alternative languages or user opt-in. Under the policy rule for language/locale constraints, this is a natural-language policy issue unless the file clearly documents a justified region-specific scope, which it does not.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The entire skill content and user-facing instructions are written exclusively in Chinese, with no indication that users may choose another language or locale. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

This markdown file contains user-facing instructional content exclusively in Chinese, and there is no indication that users may select another language or that the locale restriction is required for a specific regional use case. Under the language/locale policy check, this can be considered a forced language choice without user opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The entire document is written as Chinese-only guidance and does not indicate any option for another language or any user opt-in to a locale restriction. Under the stated policy, forcing a specific language without user choice can be a natural-language policy violation unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The prompt explicitly fixes the target readers as native English speakers and requires the model to produce language matching modern young Western slang. This imposes a specific language/locale preference with no user choice or opt-in, which matches the language/locale policy concern.

Content

No source excerpt is available for this finding.