ClawHub
Back to skill

Security audit

requirement-pre-review

Security checks across malware telemetry and agentic risk

Overview

This is a single-file requirement-review skill with no executable code, installer, persistence, credential use, or hidden data flow.

Install this if you want a Chinese-language PRD or requirements pre-review checklist. Consider narrowing the trigger wording or requiring explicit user intent if accidental activation on generic review requests would be disruptive.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
93% confidence
Finding
The trigger conditions are intentionally broad enough to activate on many generic 'review/check' style requests, including cases where the user may not actually want this specific skill. In an agent environment, over-broad activation can cause incorrect routing, unnecessary disclosure of user content to the skill, and degraded reliability because the wrong workflow is invoked by default.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The skill hard-codes Chinese-language behavior and output format without checking the user's preferred language. While not a classic security flaw, it is a policy and safety issue because it can override user intent, reduce comprehension, and increase the chance that users misunderstand important review results or follow-up questions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal