ClawHub
Back to skill

Security audit

Medical Search

Security checks across malware telemetry and agentic risk

Overview

This skill has a legitimate medical-search purpose, but it sends sensitive medication questions to a hardcoded plain-HTTP search server and optionally runs an unreviewed local script.

Review before installing. Use only if you are comfortable sending medical search terms to the listed HTTP IP address, or replace it with a trusted HTTPS search provider. Verify any optional local DDInter script before allowing the agent to run it, and treat outputs as informational rather than medical advice.

SkillSpector

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

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill instructs the agent to execute a local Python script from an external project path outside the skill boundary. This expands capability from simple search into arbitrary local code execution of a dependency that may be modified, missing, unreviewed, or malicious, creating a trust-boundary violation and increasing supply-chain risk.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.