Release Readiness Checker

Security checks across malware telemetry and agentic risk

Overview

This is a transparent release-checklist skill with mostly read-only checks, but it overstates test verification and should not be treated as a real release gate by itself.

Safe to install as an advisory checklist. Do not rely on it alone to approve a release: run the actual test suite and CI separately, review any secret-scan matches before sharing reports, and treat npm/GitHub CLI checks as networked operations using your local credentials.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
The skill description claims it will verify that tests pass, but the implementation only detects whether test infrastructure exists and explicitly avoids executing tests. This creates a dangerous trust gap: users may approve a release believing test results were validated when no such verification occurred.

Intent-Code Divergence

High
Confidence
99% confidence
Finding
The 'Tests Pass' section confirms the skill does not run tests automatically, yet the surrounding documentation and release verdict logic present this as a readiness check. In a release-gating context, this can lead to false GO decisions and shipment of untested or broken code.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal