ClawHub
Back to skill

Security audit

latent-space-engineering

Security checks across malware telemetry and agentic risk

Overview

This skill is a markdown-only prompt-writing guide with no hidden execution, persistence, credential use, or data-moving behavior found.

This skill appears safe to install as a prompt-writing aid. Be aware it may activate for some general instruction-writing or review-prompt tasks, and only include exemplar code or prose in style-transfer prompts when you are comfortable placing that material in the agent context.

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 (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger phrase "composing dispatch prompts or writing skill instructions for parallel review agents" is a long natural-language description rather than a specific activation phrase, and it lacks clear boundaries for when the skill should or should not activate. Because parts of it overlap with common authoring or review-related work, it could cause unintended invocation in broader prompting or documentation contexts.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.