Security audit

Private Advisor

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only Chinese philosophy advisor with no executable behavior, but it needs review because it covers depression, crisis-like distress, and broad high-stakes topics without clear safety boundaries.

Install only if you want a Chinese-language reflective philosophy advisor. Do not rely on it as therapy, crisis support, legal advice, financial advice, or current geopolitical guidance; it should add explicit crisis escalation, professional-care boundaries, and exclusions for high-stakes domains before broad deployment.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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 (32)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The file targets users expressing failure, hopelessness, and inability to continue, yet provides only inspirational advice without any limitation notice, safety boundary, or escalation path for sensitive mental-health scenarios. In a philosophy/advice skill, this is dangerous because distressed users may interpret the content as sufficient support, even when they may need urgent professional or crisis intervention.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill declares an extremely broad scope ('any real-world confusion' and full coverage across personal, social, political, military, ethics, history, and economics), which can cause over-activation and make the agent respond in domains where specialized safeguards are needed. In context, this is more dangerous because it includes sensitive areas like mental health, conflict, and public affairs, increasing the chance of low-context advice being treated as authoritative.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill explicitly targets emotionally sensitive issues such as anxiety, anger, emotional distress, setbacks, and relationship problems, but provides no warning that it is not a therapist, counselor, or crisis service, and gives no instructions for when to seek qualified help. This is dangerous because users may rely on philosophical reflection in situations involving depression, self-harm risk, abuse, or acute mental health crises where supportive redirection and professional resources are necessary.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 82% confidence
Finding: All instructions and content in the file are written only in Chinese, with no indication that the user can choose another language or that the skill is intentionally limited to Chinese-speaking users. Under the policy, a fixed language without user opt-in can be a locale/language policy violation unless clearly justified.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The trigger list includes broad phrases like '生活好难' and '坚持不下去了', which commonly appear in conversations involving acute emotional distress or possible self-harm risk. In this skill’s context, such broad matching can route vulnerable users into generic philosophical reframing instead of safer, crisis-aware handling, increasing the chance of harmful or delayed support.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger list is very broad and includes generic phrases such as '怎么办' and general anxiety/uncertainty themes that can appear in many ordinary conversations. This creates a real risk of unintended invocation, causing the agent to apply this philosophy-heavy skill in contexts where the user may instead need domain-specific, safety-sensitive, or more precise assistance.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation phrases are very broad, everyday expressions such as indecision and hesitation, so the skill may trigger in many normal conversations where the user did not explicitly request this philosophy-oriented advisor. This can cause unintended routing, overshadow more appropriate skills, and inject advice into sensitive contexts without clear user intent.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This file provides mental-health coping guidance for anxiety, depression, insomnia, and emotional instability, but it includes no warning that the content is not a substitute for professional care and no escalation path for crisis situations such as self-harm, suicidality, or severe deterioration. In this context, users in acute distress may rely on philosophical reassurance and simple self-help steps instead of seeking urgent support, which can delay intervention and increase harm.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The markdown specifies that the skill should be used whenever a user asks about broad topics such as ethical issues of AI, privacy and security balance, or the future of humanity. These are expansive, common conversational themes without clear boundaries or exclusion conditions, which could cause unintended invocation in general-purpose discussions.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrases are very broad emotional and everyday expressions such as '快乐不起来' and '焦虑、不满足', which can match ordinary conversation outside the intended scope. This can cause unintended invocation of the skill in sensitive contexts, leading the agent to give generalized philosophical advice when a more appropriate, domain-specific, or safety-aware response is needed.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The file content is entirely in Chinese and does not offer any indication that users can choose another language or that the skill is intentionally restricted to a Chinese-speaking audience. This can create accessibility and usability problems, and in multilingual deployments it may cause users to misunderstand responses or be excluded from using the skill effectively.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The file is entirely in Chinese and does not provide a language-selection mechanism or document that the skill is intentionally restricted to Chinese-speaking users. In multilingual environments, this can lead to misrouting, user misunderstanding, or partial/inaccurate application of the skill, which weakens reliability and may cause users to miss important nuance in sensitive geopolitical analysis.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The invocation criteria are very broad, covering military conflict, diplomacy, ethics, and general strategic analysis without clear scoping constraints. This can cause the skill to trigger for a wide range of ordinary geopolitical queries, increasing the chance that users receive this specialized framing when it was not intended and potentially biasing responses toward conflict-analysis content.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The file explicitly targets users expressing emptiness, lack of meaning, and existential distress, which are mental-health-adjacent states, but it provides only philosophical content and no safety warning, crisis check, or escalation guidance. In this context, a vulnerable user could receive abstract advice instead of support-seeking guidance, increasing the risk of harm if the content is used during acute distress.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The file explicitly covers sensitive mental health scenarios such as anxiety, insomnia, and obsessive thinking, yet provides no disclaimer that the content is not a substitute for professional care and no escalation guidance for crisis situations. In a user-facing advisory skill, this can cause users to rely on generic self-help content when they may need qualified psychological or medical support, increasing the risk of delayed care or harmful advice in high-risk contexts.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 85% confidence
Finding: The file is entirely in Chinese with no indication that the user requested Chinese or opted into that language. In an agent workflow, this can lead to the wrong skill being selected or its guidance being surfaced in a language the user may not understand, increasing the risk of miscommunication in sensitive interpersonal advice scenarios.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation examples are broad and open-ended, covering many common interpersonal situations without clear boundaries for when this reference should or should not be used. In an agent system, this can cause over-triggering, making the skill apply philosophical advice in contexts that may require specialized handling, such as mental health crises, abuse, harassment, or workplace/legal escalation.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The activation phrases are very broad, common expressions about motivation, procrastination, and self-discipline, so the skill may trigger in many ordinary conversations where the user did not explicitly ask for a philosophy-based intervention. This can cause unintended invocation, relevance errors, and override or crowd out better-matched skills, especially because the prompts are generic everyday language rather than narrowly scoped intent markers.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The trigger list is overly broad and can activate on many ordinary discussions about social issues, law, policy, or ethics. This increases the chance the skill will intercept queries outside its intended scope and steer responses using a fixed ideological/philosophical framing, which is a scope-control and policy-routing weakness rather than direct code execution risk.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This markdown skill file presents its title, instructions, and operational guidance exclusively in Chinese, and the usage instructions do not indicate that users may opt into another language. Under the policy criteria, forcing a specific language without user choice can be a natural-language policy violation unless the locale restriction is explicitly justified.
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 description and prescribed output format are written only in Chinese, with no indication that the user may choose another language. Under the language/locale policy rule, this is a natural-language policy concern unless the skill explicitly offers language choice or documents a justified locale restriction.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The natural-language content and invocation guidance are fully Chinese-language, which can amount to forcing a specific language on users if no opt-in or documented regional scope is provided. The file does not indicate that the skill is China-specific or that users may choose another language.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 83% confidence
Finding: This markdown file is entirely written in Chinese and instructs users to contact the author via WeChat, with no indication that other languages are supported or that the language choice is optional. Under the policy rule for language/locale constraints, this can be read as imposing a specific language without user opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: All visible instructional and user-facing content in the file is Chinese, and the document does not indicate that the user can choose another language or that the skill is intentionally limited to Chinese-speaking contexts. 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: 93% confidence
Finding: All user-facing content in the file is written exclusively in Chinese, and there is no indication that the skill is region-specific or that the user can opt into this language. That creates a natural-language policy concern because the skill implicitly enforces a single language without user choice or justification.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.