ClawHub
Back to skill

Security audit

Watcha Finder

Security checks across malware telemetry and agentic risk

Overview

This instruction-only product recommendation skill uses watcha.cn and web search, with no hidden code, credential access, persistence, or destructive behavior found.

Reasonable to install for non-sensitive AI product research. Be aware that searches may go to watcha.cn and general web search providers, and recommendations may reflect watcha.cn coverage, Chinese-language search expansion, and community-review bias.

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 (2)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill’s trigger conditions are broad enough to activate on many generic AI-tool queries, which can cause unnecessary routing to this skill even when the user did not ask for watcha.cn specifically. That increases the chance of over-collection of user intent, biased recommendations anchored to a single platform, and unintended external API use without clear user expectation.

Natural-Language Policy Violations

Medium
Confidence
81% confidence
Finding
The workflow instructs the agent to try Chinese search terms as part of its default search strategy, which can steer behavior toward Chinese-language querying regardless of the user’s language or regional preference. This may produce irrelevant or regionally mismatched results, leak user intent to an external service in a different linguistic context, and reduce user control over how their query is processed.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.