短链接 - 即刻数据

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward short-link tool that uses the disclosed JikeAPI service, with privacy considerations around sending URLs and an API key to that provider.

Install only if you are comfortable sending link data and your JikeAPI key to jikeapi.cn. Avoid using it with private internal URLs, secrets, or sensitive identifiers, and verify JIKE_API_BASE_URL is not set to an unexpected host.

SkillSpector

By NVIDIA
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 (3)

Vague Triggers

Medium
Confidence
76% confidence
Finding
The description includes broad natural-language trigger examples like '帮我生成一个短链接' and '查一下这个短链访问次数', which can overlap with ordinary conversation and cause unintended invocation. In this skill, accidental activation matters because invocation may send URLs and analytics queries to a third-party service and may create real short links as a write operation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
Although the markdown names the provider, it does not clearly warn users that submitted original URLs, short links, access statistics queries, and the app key are sent to a third-party API service. This creates a privacy and transparency issue, especially for restoration and statistics features that may expose internal or sensitive links and usage metadata to an external operator.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill sends the user's target/link values and the AppKey to a remote third-party API, but the runtime behavior does not prominently warn the user that their provided URLs and credential-backed request will leave the local environment. In an agent setting, this can create unintended data disclosure, especially if users submit sensitive internal URLs or assume processing is local.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal