ClawHub
Back to skill

Security audit

safe-refactor

Security checks across malware telemetry and agentic risk

Overview

This skill gives cautious instructions for behavior-preserving code refactors and does not add hidden execution, data access, persistence, or credential handling.

Install this if you want an agent to perform cautious refactors with tests and verification. Be aware that vague requests like "clean this up" may activate the refactor workflow, so use explicit wording when you want feature changes or bug fixes instead.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrase "clean this up" is broad and can match many user requests that do not clearly mean a strict behavior-preserving refactor. That can cause the skill to activate in contexts where the user intended bug fixing, feature changes, or general code improvement, leading the agent to take the wrong workflow and make unauthorized or misleading changes.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The phrase "do it" is highly ambiguous and lacks a clear reference to the safe-refactor workflow. In multi-step agent interactions, this can cause unintended activation based on stale or partial context, making the agent perform refactoring actions the user did not specifically authorize.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.