Intercom GenAIgrp

Security checks across malware telemetry and agentic risk

Overview

This skill transparently routes explicit user-written messages to another OpenClaw session and shows no hidden execution, persistence, credential access, or destructive behavior.

Install only if you want explicit !target messages routed between OpenClaw sessions. Verify the target session key before sending, and do not include secrets or private data unless the receiving session is trusted and intended to see it.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill instructs the agent to invoke an inter-session messaging capability (`sessions_send`), which is effectively a communication/network-like action, but no explicit permission model or authorization constraints are declared in the skill. That gap can enable unintended message routing, spoofed-looking internal communications, or abuse of agent-to-agent channels if the runtime does not enforce strict access controls outside the skill.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal