ClawHub
Back to skill

Security audit

loop-constructor-codex

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed design/runbook generator that writes bounded `.loop/` artifacts and does not hide execution, credential use, or exfiltration behavior.

Install this only if you want the skill to create loop design files in your project, usually under `.loop/`. Review generated runbooks before running any suggested `codex exec`, git worktree, or check commands, because those later operator steps can modify the target repository even though the skill itself is design-only.

SkillSpector

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Low
Confidence
84% confidence
Finding
The skill explicitly instructs persisting generated artifacts into the target project's `.loop/` directory, which is a filesystem modification. While the modification is limited to design/runbook artifacts and the skill repeatedly says not to execute the designed loop, it still causes writes to user files without an explicit up-front warning or confirmation step, which can lead to unintended repository changes or pollution of a working tree.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.