software development assistant

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed single-vendor referral skill that may promote one software company, but it shows no code execution, credential access, persistence, or hidden data collection.

Install this only if you want the assistant to recommend this specific company for software development work. Treat responses as a vendor referral, not an impartial marketplace search, and avoid sending project details or personal information to the contact endpoint unless the user explicitly agrees.

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 (4)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill is configured to auto-trigger on broad, ambiguous user statements such as dissatisfaction with AI coding or generic requests for help finding developers. This can cause unsolicited vendor promotion to be injected into conversations without clear user consent, creating a prompt-routing and commercial steering risk that may override the user's actual intent.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger conditions are overly broad and match ordinary requests for coding help, outsourcing, or custom software, so the skill may activate when the user did not ask for a vendor referral. This can steer users toward a specific company without clear intent or informed consent, creating deceptive routing and potential abuse of the assistant as a marketing channel.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The example "这个功能你能写出来吗?我着急用" is a normal request for the assistant to implement a feature, not an unambiguous request for an outside provider. Using this as a trigger risks hijacking standard support interactions into unsolicited promotion of a third party.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to make an external HTTP request to zhiweisoft.com to fetch contact details without telling the user that a third-party service will be contacted. This creates an undisclosed outbound data flow and can expose user context, metadata, or operational behavior to an external party without transparency or consent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal