Annature

Security checks across malware telemetry and agentic risk

Overview

This Annature skill is a coherent integration, but it gives an agent broad ability to change signing and account data without clear approval boundaries.

Review before installing. Use a least-privileged Annature account, connect only the intended workspace, consider pinning or verifying the Membrane CLI version, and require the agent to show the exact action, target record, HTTP method, and JSON payload before any create, update, delete, or direct proxy request.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill explicitly documents create, update, and proxy request capabilities, including arbitrary direct API access, without guardrails about user confirmation, scope limitation, or sensitivity of remote side effects. In an agent setting, this increases the chance of unintended modification of Annature documents, recipients, accounts, or workflows, especially if a prompt is ambiguous or adversarial.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal