Qdrant

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate Qdrant integration, but it gives an agent broad authenticated ability to change or delete live database data without clear approval safeguards.

Install only if you are comfortable letting an agent act through your Membrane-connected Qdrant account. Prefer a least-privilege Qdrant credential, review any direct API request before it runs, and require explicit confirmation for create, update, delete, snapshot, lock, or service-level operations.

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: 88% confidence
Finding: The skill explicitly instructs the agent to use a generic proxy interface capable of GET, POST, PUT, PATCH, and DELETE against the Qdrant API, but it does not require confirmation or caution for destructive operations. In an agent setting, this increases the chance that a model will perform state-changing or deletion requests on collections, points, or snapshots without sufficient user awareness, especially because Qdrant is a live data store.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal