ClawHub

WoClaw

Security checks across malware telemetry and agentic risk

Overview

This skill is a documented connector for a user-configured WoClaw Hub, with expected shared-memory and network-token risks but no hidden or unrelated behavior in the submitted artifacts.

Install only if you trust the WoClaw Hub and any hub package or container you deploy. Do not put secrets, private prompts, credentials, personal data, or regulated data into topics or shared memory unless the hub is properly secured; prefer WSS/HTTPS or trusted private networks, keep WOCLAW_TOKEN out of committed config, logs, screenshots, and shell history, and rotate/scope tokens where possible.

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 (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README promotes shared memory and multi-agent topic messaging as core features but does not warn that any data written may be visible to other connected agents, retained by the hub, or replayed through version/history features. In an agent ecosystem, this can lead to accidental disclosure of prompts, secrets, internal state, or sensitive task outputs because users may assume memory is local or ephemeral when it is actually shared infrastructure.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The setup instructions direct users to configure a remote hub URL and authentication token, but they do not explain credential sensitivity, transport security expectations, or the risks of sending agent data to a network service. This can cause users to expose tokens in shell history/config files or send sensitive agent traffic over insecure or untrusted connections, especially since the examples emphasize ws/http connectivity without security guidance.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal