Clickhouse

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate ClickHouse integration, but it gives an agent broad authenticated database access without clear confirmation guardrails for write or delete operations.

Install only if you trust Membrane and the npm CLI source. Use a least-privileged or read-only ClickHouse account where possible, and require explicit review before allowing the agent to run raw proxy requests, schema changes, writes, or deletes against production data.

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: 94% confidence
Finding: The skill explicitly documents a generic proxy mechanism that can send arbitrary requests, including destructive HTTP methods, to the ClickHouse API without requiring an explicit confirmation or warning about network transmission and state-changing operations. In a database context, this increases the chance of unintended data modification, deletion, or exfiltration if an agent uses the proxy path too freely.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal