native-mcp

Connect to external MCP servers and expose their tools as native Hermes Agent capabilities.

641|104|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/eastreams/loong --skill native-mcp-eastreams
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: native-mcp
Source: https://github.com/eastreams/loong/tree/main/skills/native-mcp
Command: npx skills add https://github.com/eastreams/loong --skill native-mcp-eastreams

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The native MCP client eliminates the need for a separate bridge by automatically discovering external MCP servers and exposing their tools as native Hermes Agent commands, enabling seamless access to additional capabilities.

Core Features & Use Cases

  • Automatic discovery of MCP tools during agent startup.
  • Supports both stdio (command‑based) and HTTP transports.
  • Secure environment variable filtering and credential redaction.
  • Persistent connections with automatic reconnection and graceful shutdown.
  • Ideal for integrating filesystem access, GitHub management, custom APIs, or any MCP‑compatible service into agent workflows.

Quick Start

Ask the agent to retrieve the current time using the MCP time tool.

Frequently Asked Questions about native-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I connect an external MCP server to my agent for tool discovery?

To connect an external MCP server to your agent, the native MCP client automatically discovers available tools during startup and exposes them as native capabilities, eliminating the need for a separate bridge.

Does MCP integration support both stdio and HTTP transport connections?

Yes, MCP integration supports both stdio command-based and HTTP transports. It maintains persistent connections with automatic reconnection and graceful shutdown for reliable agent workflows.

How is secure environment handling managed during MCP agent integration?

Secure environment handling during MCP agent integration is managed through environment variable filtering and credential redaction, ensuring sensitive data is protected when connecting to external services.

Can I use MCP tools for filesystem access and GitHub management within my agent?

Yes, you can use MCP tools for filesystem access and GitHub management. The native MCP client integrates any MCP-compatible service, including custom APIs, directly into your agent workflows.

Why do I need a native MCP client instead of a separate bridge for agent integration?

You need a native MCP client instead of a separate bridge because it automatically discovers external MCP servers and exposes their tools as native agent commands, enabling seamless access to additional capabilities.

How do I test MCP tool discovery after configuring my agent?

To test MCP tool discovery after configuring your agent, ask the agent to retrieve the current time using the MCP time tool. This verifies the automatic discovery and native command exposure.