Lizhi

Security checks across malware telemetry and agentic risk

Overview

This skill is a lightweight guide for summarizing public Lizhi FM pages and does not include code execution, credentials, persistence, downloads, or hidden data handling.

Install this only if you want public Lizhi FM page summaries and ranking/search analysis. Keep usage light, user-directed, and compliant with platform rules; avoid using it for bulk scraping, downloads, account-only content, or attempts to bypass site controls.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 (1)

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The trigger description is overly broad: "当需要访问或自动化荔枝相关内容时调用" can match many generic requests involving Lizhi/FM-related browsing or automation, increasing the chance this skill is invoked outside its intended narrow scope. Overbroad invocation can cause unintended web access or data collection behavior, especially in agent systems that select skills from natural-language descriptions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal