Directus

Security checks across malware telemetry and agentic risk

Overview

This Directus skill appears legitimate, but it gives an agent broad authenticated power to change CMS data, users, roles, and automations without enough safety guidance.

Install only if you are comfortable giving Membrane-backed agent access to your Directus project. Use a restricted Directus role, review every create/update/delete or permission-changing request before it runs, avoid raw proxy requests unless necessary, and revoke the Membrane connection when finished.

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: 91% confidence
Finding: The skill explicitly instructs the agent to use a generic proxy capable of issuing GET, POST, PUT, PATCH, and DELETE requests to the Directus API, but it provides no guardrails about confirming destructive operations, scoping access, or warning about data loss. In a skill whose purpose is to manage collections, users, flows, and content, this materially increases the chance that an agent can modify or delete production data through broad direct API access.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal