ClawHub
Back to skill

Security audit

Grupr

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-built for Grupr chat integration, but it needs user review because it asks users to handle a live browser session JWT in an unsafe, under-warned way.

Install only if you are comfortable running a background chat bridge that reads selected Grupr conversations, sends messages to your local OpenClaw agent, and posts replies as that agent. Treat the Grupr JWT as highly sensitive, avoid pasting it into shared terminals or logs, clear shell history if needed, and prefer a dedicated Grupr/OpenClaw agent for conversations that do not contain secrets.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README instructs users to extract a live JWT from browser DevTools cookies and pass it on the command line, but provides no warning about the sensitivity of that credential or safer handling guidance. JWTs and shell arguments can be exposed through shell history, process listings, terminal scrollback, screenshots, or logs, which could let an attacker reuse the token to impersonate the user or mint agent tokens.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The description says the skill adds an agent to a Grupr conversation, but it does not prominently warn that conversation content is continuously sent to a local OpenClaw gateway and that replies are automatically posted back as the agent. This can cause unintended disclosure of private chat content and autonomous actions under the agent identity without fully informed user consent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The recovery guidance instructs users to retrieve a session JWT from browser DevTools cookies, which is a highly sensitive credential handling workflow. Documenting token extraction without strong warnings normalizes unsafe secret exfiltration practices and increases the chance the JWT is mishandled, reused insecurely, or exposed in shell history and logs.

VirusTotal

No VirusTotal findings

View on VirusTotal