ClawHub

PRD Generator - 专业产品需求文档生成器

Security checks across malware telemetry and agentic risk

Overview

This is a document-writing helper for PRDs, with no executable code or requests for credentials, local data, network access, persistence, or privileged actions.

Reasonable to install for drafting PRDs. Because PRD work may involve sensitive product plans, avoid pasting secrets, customer data, confidential roadmap details, or proprietary implementation details unless your workspace policy permits it. If the skill activates during general product brainstorming, ask the agent to stay in discussion mode or explicitly request PRD output only when needed.

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
83% confidence
Finding
The activation criteria are broad enough that the skill could trigger during general product discussions, causing the agent to switch into PRD-generation mode without explicit user intent. In an agent system, over-broad routing is dangerous because it can override user expectations, produce irrelevant structured outputs, and increase the chance of mishandling sensitive business context under the wrong workflow.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The repeated broad trigger phrases reinforce permissive activation without sufficient guardrails, making unintended invocation more likely across ordinary product-work conversations. This is more dangerous in context because the skill is designed to generate long, structured artifacts, so accidental activation can materially derail the session and expose more internal context than necessary to a templated workflow.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal