native-mcp

Connects MCP servers via stdio or HTTP and registers their tools automatically.

Updated May 4, 2026
One-click install
npx skills add https://github.com/Junkfooooood/hermes-governance --skill native-mcp-junkfooooood
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: native-mcp
Source: https://github.com/Junkfooooood/hermes-governance/tree/main/skills/mcp/native-mcp
Command: npx skills add https://github.com/Junkfooooood/hermes-governance --skill native-mcp-junkfooooood

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mcp.

What problem does it solve? Integrating external MCP servers into an agent normally requires custom bridge code and manual tool wiring. This Skill configures Hermes Agent's built-in MCP client so server tools are discovered at startup and registered as first-class tools alongside built-in ones. ## Core Features & Use Cases - Stdio and HTTP Transports: Run local MCP servers via npx/uvx commands or connect to remote StreamableHTTP endpoints with custom headers. - Automatic Tool Discovery: On startup, tools are discovered and registered with the mcp_{server}_{tool} naming pattern, then injected into all platform toolsets. - Security Controls: Environment variable filtering prevents credential leakage to subprocesses, and credential-like patterns are redacted from error messages. - Use Case: Add a GitHub MCP server with a personal access token in ~/.hermes/config.yaml, restart the agent, and immediately call tools like mcp_github_list_issues in any conversation. ## Quick Start Add an MCP server entry under the mcp_servers key in ~/.hermes/config.yaml and restart the agent to make its tools available.

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 MCP server to Hermes Agent?

Add an entry under the mcp_servers key in ~/.hermes/config.yaml with either a command (stdio) or url (HTTP), then restart the agent. Tools are discovered at startup and registered with the mcp_{server}_{tool} prefix.

How do I run npx or uvx MCP servers locally?

Use the stdio transport by setting command to npx or uvx and passing the package name in args, for example args: ["-y", "@modelcontextprotocol/server-filesystem"]. Node.js is required for npx servers and uv for uvx servers.

Does the MCP client support remote HTTP servers?

Yes, HTTP and StreamableHTTP transports are supported by setting a url and optional headers such as Authorization tokens. Your installed mcp package must include mcp.client.streamable_http, otherwise the server fails with an ImportError.

Why are my MCP tools not appearing after configuration?

Check that servers are listed under mcp_servers (not mcp or servers), YAML indentation is correct, and the mcp package is installed. Tool names follow the mcp_{server}_{tool} pattern, so search for that prefix in startup logs.

Are my API keys safe when passed to MCP servers?

Stdio subprocesses receive only a filtered baseline environment (PATH, HOME, and similar), not your full shell environment. Secrets must be explicitly declared via the env config key, and credential patterns are redacted from error messages shown to the LLM.

What is the difference between native MCP and mcporter?

The native MCP client provides persistent, auto-discovered tools configured in the agent config and available in every conversation. mcporter is for ad-hoc, one-off MCP tool calls from the terminal without any configuration, and both can be used simultaneously.