Post-Development Verification

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:34 PM UTC
• Version:
  1.0.1

Overview

This is a testing/verification skill with powerful but disclosed test-environment actions and no evidence of hidden code, exfiltration, or deceptive behavior.

Install this only if you want an agent to perform real verification work. Before allowing execution, review the Phase 0 environment report and confirm that all services, URLs, databases, credentials, cleanup commands, and generated tests target disposable test resources, not production or staging systems.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The skill is configured to auto-trigger on very broad phrases like 'run tests', 'verify', 'does this work?', and pre-merge situations, which can overlap with ordinary conversation and cause unintended activation. In this skill's context, unintended activation is meaningful because the workflow explicitly permits starting services, running migrations, seeding data, and making network calls, increasing the chance of unreviewed side effects in the wrong project or environment.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The cleanup guidance includes destructive actions like dropping and recreating a test database or deleting all resources created during a run, but it does not require an explicit warning, environment scoping, or safeguards to prevent these steps from being used against non-test systems. In a skill that emphasizes real execution and automated verification, operators or downstream agents may copy these patterns into production-adjacent contexts, causing unintended data loss.

VirusTotal

64/64 vendors flagged this skill as clean.