ClawHub

Mcp Config

Security checks across malware telemetry and agentic risk

Overview

This skill is mostly a coherent MCP setup guide, but it includes unsafe database configuration examples that could give an agent broad database access without adequate warning.

Review carefully before installing. Use this skill only for MCP configuration tasks, and avoid copying the PostgreSQL examples as-is; prefer read-only or least-privilege database roles, keep real credentials out of shared config files, and confirm any global or project MCP changes before allowing an agent to apply them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The use-when guidance lists examples but does not define firm activation boundaries or non-goals, so an orchestrator may invoke this skill for loosely related support requests. In a skill that can influence MCP server configuration and troubleshooting steps, ambiguous routing increases the chance of unnecessary or unsafe configuration changes.

Vague Triggers

Low
Confidence
82% confidence
Finding
The use-when guidance lists examples but does not define firm activation boundaries or non-goals, so an orchestrator may invoke this skill for loosely related support requests. In a skill that can influence MCP server configuration and troubleshooting steps, ambiguous routing increases the chance of unnecessary or unsafe configuration changes.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The catalog provides a PostgreSQL MCP example using `--access-mode=unrestricted` and a live `DATABASE_URI` pattern without any warning about the security implications. In a configuration/diagnostics skill, this can normalize unsafe defaults and lead users to connect an agent to a database with broad read/write capability, increasing the risk of destructive queries, data exfiltration, or accidental modification.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation includes a concrete database connection string containing embedded credentials (`postgresql://user:pass@host:5432/db`) without any warning that secrets must not be hardcoded. In a configuration skill, users are likely to copy examples directly into real MCP configs, which can normalize insecure secret handling and lead to credential exposure through files, repos, logs, or screenshots.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal