fullstack-dev-engineer

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-style full-stack development helper, with some examples users should review carefully before copying into real apps or deployments.

Safe to install as a reference/code-generation helper. Before using its generated auth or deployment examples, replace placeholders, avoid storing long-lived tokens in browser storage, use least-privilege CI/CD secrets, verify Kubernetes context and namespace, and require human review for production changes.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Intent-Code Divergence

Medium

Confidence: 98% confidence
Finding: The guide contains contradictory security guidance: the Pinia example persists an auth token in localStorage, while a later security section says not to store sensitive tokens in frontend storage. In a frontend engineering guide, readers may copy the earlier code directly, and localStorage-stored tokens are accessible to JavaScript and therefore exposed if XSS occurs, increasing account/session compromise risk.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill description is extremely broad, covering architecture, tech selection, frontend, backend, deployment, code generation, and best practices. In an agent-routing context, such a generic trigger can cause the skill to activate for a wide range of ordinary developer prompts, increasing the chance of unintended delegation, prompt-scope capture, or overshadowing of more specialized and safer skills.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger condition states the skill should be used whenever the user needs system design, tech selection, frontend/backend development, deployment, code generation, or best-practice advice, which is effectively most software engineering requests. This ambiguity can make the orchestration layer over-select this skill, reducing predictability and potentially exposing users to irrelevant or over-privileged behavior.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal