Lanxin App Articles

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:42 PM UTC
• Version:
  1.0.0

Overview

This skill is a narrow Lanxin article-card JSON formatter, but its broad triggers may make it activate when the user only mentions articles or image-text content.

Install this only if you want the assistant to emit raw Lanxin appArticles JSON. Review generated titles, URLs, image links, and recipients before using the payload, and be aware that generic requests involving articles or image-text content may trigger JSON-only output.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

High

• Confidence: 93% confidence
• Finding: The trigger list includes broad everyday terms like '图文' and '文章', which can cause the skill to activate in many unrelated conversations. In an agent setting, this can unexpectedly force the assistant into emitting raw JSON-only output, disrupting normal behavior and potentially causing unintended downstream actions or message dispatch formatting.

Overly Broad Trigger

Low

• Category: Trigger Abuse
• Confidence: 88% confidence
• Finding: The trigger '蓝信' is overly short and likely to match benign mentions of the platform, causing accidental activation. In context, accidental activation is somewhat more dangerous because the skill instructs the model to output only raw JSON and forbids normal explanatory text, which can override safer conversational behavior.

Overly Broad Trigger

Low

• Category: Trigger Abuse
• Confidence: 96% confidence
• Finding: The trigger '图文' is extremely generic and could match many unrelated requests about articles, image-text content, or document formatting. Because the skill then demands JSON-only output, this broad match can hijack unrelated interactions and cause malformed workflow routing or unintended payload generation.

Overly Broad Trigger

Low

• Category: Trigger Abuse
• Confidence: 95% confidence
• Finding: The trigger '文章' is a common everyday word and is highly likely to produce false activations on unrelated writing or reading tasks. In this skill, that risk is amplified by rigid instructions to emit only JSON, which can derail ordinary user requests and potentially trigger unintended integration behavior.

VirusTotal

66/66 vendors flagged this skill as clean.