Leexi

Security checks across malware telemetry and agentic risk

Overview

This Leexi integration is coherent, but it gives an agent broad authenticated ability to read, create, modify, and delete business communication data without enough guardrails.

Install only if you trust Membrane as the bridge to your Leexi workspace and are comfortable granting delegated access. Use a least-privileged Leexi account where possible, confirm every create/update/delete action immediately before it runs, and treat raw proxy requests as high-risk because they can reach endpoints beyond the listed actions.

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

Low

Confidence: 83% confidence
Finding: The skill advertises a delete action without stating that destructive operations require explicit user confirmation. In an agentic context, this can lead to accidental deletion of meeting events if the model infers intent too aggressively or acts on ambiguous prompts.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal