Security audit

mcp-config

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only MCP configuration helper, but its PostgreSQL examples encourage unrestricted database access and inline credentials without enough safety guidance.

Review the PostgreSQL examples before using them. Prefer read-only or least-privilege database credentials, avoid committing secrets to project `.mcp.json`, and keep sensitive MCP servers in local or user scope only when that is intentional.

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 (2)

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The postgres example explicitly configures the server with `--access-mode=unrestricted`, which encourages broad database access without any warning, scoping guidance, or least-privilege safeguards. In a catalog meant for copy/paste configuration, users may adopt this setting verbatim and expose modification or destructive access to production data through the MCP server.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The example embeds a full PostgreSQL connection URI with inline username and password directly in documentation. Even if presented as a sample, this normalizes insecure secret handling and increases the chance users will copy real credentials into config files, shell history, repositories, or logs, especially in a skill specifically meant to help users configure MCP servers.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.