ClawHub

Sage 300

Security checks across malware telemetry and agentic risk

Overview

This Sage 300 skill appears legitimate, but it gives an agent broad ERP access that could change sensitive business records without clear built-in safeguards.

Install only if you trust Membrane and intentionally want agent-mediated Sage 300 access. Use a least-privilege or test account, prefer read-only discovery and typed Membrane actions first, and require explicit user confirmation before any create, update, delete, payment, purchase, approval, journal-entry, workflow, user/role, or raw proxy request.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description is broad enough to match many generic ERP or business-data requests, which can cause over-invocation of a high-privilege integration. Because the skill can access and potentially modify finance, inventory, customer, and workflow data, accidental activation increases the chance of unintended sensitive-data access or destructive operations.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documentation describes how to run actions and proxy requests against Sage 300 but does not warn that the integration can create, update, delete, or otherwise alter business-critical records. In an ERP context, missing warnings materially increases the risk of accidental financial, inventory, payroll, or customer-data changes by users or downstream agents.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal