Characterization Test Writing

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:43 PM UTC
• Version:
  1.0.0

Overview

This is a markdown-only developer skill for writing regression-safety tests around legacy code, and its requested code edits and test runs match that purpose.

Install this if you want the agent to help write characterization tests before changing legacy code. Review generated tests carefully because they intentionally preserve current behavior, including behavior that may later turn out to be a bug.

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 (1)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The manifest description contains many broad trigger phrases such as general requests about testing legacy code, regression safety nets, and writing tests before changes. In an agentic skill system, overly broad activation criteria can cause the skill to trigger in contexts where it is only partially relevant, increasing the chance of inappropriate tool use (Read/Edit/Bash) or workflow hijacking away from a more suitable skill. The content itself is not overtly malicious, but the breadth of matching expands the attack surface for unintended invocation.

VirusTotal

64/64 vendors flagged this skill as clean.