ClawHub
Back to skill

Security audit

Gr Competitor

Security checks across malware telemetry and agentic risk

Overview

This skill openly provides competitor-research workflows that use Actionbook to scan websites, with practical privacy and compliance cautions but no evidence of hidden or malicious behavior.

Install this only if you want an agent to use Actionbook and a local browser to fetch competitor website data. Review Actionbook and any browser extension before installing, provide explicit target URLs, avoid logged-in or restricted sites unless you intend that access, and keep bulk scans within site terms and applicable law.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrases are broad enough to match common business requests such as '分析对手' or '找对标', which can cause the skill to activate in contexts where the user did not explicitly consent to competitor scraping or third-party data collection. In this skill, unintended activation is more concerning because it can launch bulk extraction workflows, invoke external tooling, and normalize collection from multiple third-party sites without an explicit confirmation step.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The description promotes batch collection from competitor sites and use of external services, but it does not clearly warn users that requests may be sent to third-party sites, local browser tooling, and optional external providers. This missing disclosure increases the risk of users unknowingly initiating large-scale scraping, transmitting sensitive target lists or research intent to outside systems, and creating compliance or terms-of-service issues.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrases are broad enough to match common analytical requests like competitor analysis, pricing teardown, or benchmarking, which can cause the skill to activate when the user did not explicitly intend bulk external scanning. In this skill’s context, unintended invocation is more dangerous because execution can launch high-concurrency website fetching through actionbook, creating privacy, compliance, and operational side effects beyond a normal text-only response.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs installation and use of actionbook for extracting data from external sites, but it does not clearly warn users that target URLs, browsing actions, session data, or extraction activity may be transmitted to or mediated by an external tool. This is risky because users may unknowingly trigger third-party processing of sensitive competitive targets or their local browser context, especially given the skill’s bulk and concurrent scanning design.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.