other
Error
- Location
skills/xiaohongshu-90-copy/SKILL.md:50- Finding
Unsupported Medical Cases and Professional Testimony May Be Presented as Authentic
- Content
View full analysis
- Remediation
View remediation
Security audit
茶不二
Security checks for vulnerabilities and agentic risk
Overview
This is a coherent Xiaohongshu marketing skill, but it needs Review because it can create child-health promotional content using unverified clinical anecdotes and asks readers to share health data without adequate safeguards.
Install only if you are comfortable with a Chinese Xiaohongshu marketing assistant for pediatric vision-care content. Before publishing, require verified user-provided cases, confirmed credentials, human medical/legal review, and removal of any prompts asking parents to post children's ages, measurements, or reports in comments or ordinary DMs.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
Findings (2)
T09 · Insecure Skill Coding Practices
Warning
- Location
skills/xiaohongshu-90-copy/SKILL.md:91- Finding
Workflow Solicits Children's Health Reports Without Privacy or Data-Handling Safeguards
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- 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
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The file's user-facing instructions and section content are written in Chinese, and there is no indication that the skill is intentionally limited to Chinese-speaking users or a specific region. This can violate language/locale policy expectations when a skill forces one language without user opt-in or justification.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 85% confidence
- Finding
The skill title and the entire operating specification are written to require Chinese-language behavior for Xiaohongshu content generation, but the file does not offer users an opt-in or alternative language choice. Under the policy, forcing a specific language or locale without user choice is a natural-language policy violation unless clearly justified as region-specific compliance tooling.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The activation description is broad enough to trigger on generic content creation, optimization, planning, and growth-strategy requests, not just tightly scoped Xiaohongshu operations for this specific brand. That can cause unintended invocation and route unrelated user tasks into a highly specialized marketing agent that also chains other skills, increasing the chance of overscoped actions, irrelevant advice, or unnecessary access to bundled capabilities.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The README says the skill can be invoked directly 'when editing documents,' which is overly broad and can cause the agent to apply the skill during ordinary editing workflows without a narrowly scoped user request. In a skill designed to rewrite text to conceal AI-generated characteristics, broad activation increases the chance of unintended content transformation, policy evasion, or misuse in contexts where provenance and faithful editing matter.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The skill explicitly encourages users to send report data for analysis and uses examples like posting age and numbers or sending reports directly, which can solicit sensitive child health information. In the context of pediatric myopia/vision-care content, this increases the chance of collecting medical data in public comments or insecure channels without privacy guidance or minimization.
- Content
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The instructions promote direct submission of diagnostic reports and child age/vision metrics for individualized analysis, which is a form of sensitive health-data collection. Because the skill is designed to drive engagement on social platforms, the context makes this more dangerous: users may be nudged to disclose medical information in semi-public spaces where it can be retained, shared, or misused.
- Content
Missing User Warnings
Low
- Category
- Not specified by scanner
- Confidence
- 79% confidence
- Finding
This markdown skill explicitly accepts either a user's draft or a path to a draft file, then says to always produce a revised markdown file. The description does not warn whether the original file may be modified, whether a new file will be created instead, or that file-path handling can affect user data, which is a user-facing omission under the markdown warning criterion.
- Content
Static analysis
No suspicious patterns detected.