Security audit

小象中医体质辨识

Security checks for vulnerabilities and agentic risk

Overview

This is a local TCM questionnaire skill, but it should be reviewed because it collects health details and gives medicinal-food recommendations without clear consent or safety boundaries.

Install only if you are comfortable with the agent asking users health-related questions in Chinese and producing TCM-style food recommendations. The publisher should add explicit consent, privacy expectations, opt-out/skip language, and a clear medical disclaimer with red-flag and special-population guidance before broad use.

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 (5)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger conditions are broad enough to activate on generic health, wellness, or assessment requests that may fall outside the intended TCM constitution-screening scope. This can cause the agent to solicit structured health information when the user did not specifically ask for this workflow, increasing the chance of unnecessary collection of sensitive data and inappropriate guidance.

Natural-Language Policy Violations

Medium

Confidence: 88% confidence
Finding: The entire skill is written to operate in Chinese and instructs fixed Chinese questionnaire wording, with no indication that users may choose another language. Because the file does not document a justified region-specific restriction or offer opt-in language selection, it presents a language/locale policy concern.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill asks for multiple categories of health-related information, including body type, sweating, mood, sleep, and gastrointestinal symptoms, but provides no upfront notice that sensitive personal health data will be collected. Without a brief warning and consent checkpoint, users may disclose special-category data unexpectedly, creating privacy and compliance risk.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The file provides actionable health-related recommendation logic and a concrete output format for recommending foods based on constitution/symptoms, but it does not include any safety boundary, uncertainty notice, or instruction to defer to a clinician for serious, persistent, or high-risk conditions. In a medical-adjacent skill, this can lead users to treat generated food/herbal suggestions as personalized medical advice despite contraindications, comorbidities, pregnancy, medications, or emergency symptoms not being screened here.

Natural-Language Policy Violations

Low

Confidence: 88% confidence
Finding: The entire skill content, headings, and output template are written only in Chinese, with no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is explicitly documented and justified.

Static analysis

No suspicious patterns detected.