mcp-server-catalog-reference

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends the installation of various MCP servers from the NPM registry. Many of these packages are official components within the @modelcontextprotocol scope, representing the official ecosystem. It also references external transport URLs for hosted services (e.g., hosted on Glitch or vendor domains like ara.so).
  • [REMOTE_CODE_EXECUTION]: Provides instructions to download and execute remote code using the npx utility. To mitigate risks, the skill includes explicit guidance on pinning package versions and treating all third-party servers as untrusted code until verified.
  • [COMMAND_EXECUTION]: Includes shell commands for the claude CLI to add servers and modify local configuration files (e.g., claude_desktop_config.json, settings.json). These commands are standard for managing AI agent capabilities.
  • [DATA_EXFILTRATION]: Addresses the management of sensitive credentials such as database connection strings and API keys. The skill promotes security best practices by instructing the agent to use environment variables, read-only replicas, and sandbox modes for financial and infrastructure tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 10:43 PM
Security Audit — agent-trust-hub — mcp-server-catalog-reference