ClawHub
Back to skill

Security audit

Agent Communication

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent cross-agent messaging guide, with one helper script that persistently broadens OpenClaw session visibility if a user chooses to run it.

Install only if you want agents to communicate through OpenClaw sessions. Review the helper script before running it: it sets session visibility to all, which may let more agents see or interact with session context than a narrow configuration would. Keep the generated backup so you can restore the prior setting.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The script directly modifies a user security-relevant configuration file and broadens session visibility to "all" without any confirmation, validation of user intent, or discussion of the security implications. In this context, the change enables cross-agent access and is therefore more dangerous than a routine preference update because it may expose conversations or agent context more broadly than the user expects.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.