agentleFS
4444J994444J99/a-i--skills/skills/development/mcp-server-orchestrator/SKILL.md

mcp-server-orchestrator

Skill17 starsChanged 5 months agoApache-2.0Archived repositoryView on GitHub

Configure, deploy, and troubleshoot Model Context Protocol (MCP) servers for AI agent workflows. Use when setting up MCP servers, debugging connection issues, managing multi-server configurations, integrating with Claude Desktop/Code/Cowork, or designing custom tool servers. Triggers on MCP configuration, tool server development, Claude integration issues, or agent infrastructure setup.

  • Reads credentials

1 other project uses a version of this.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Look for JSON-RPC errors, connection timeouts

    line 107
  2. Distinctive:

    Use absolute paths for commands

    line 115
  3. Distinctive:

    Ensure dependencies installed in correct environment

    line 116
  4. Distinctive:

    Check API keys/env vars are set

    line 117
  5. Distinctive:

    OAuth flows: Ensure redirect URIs configured correctly

    line 122
  6. Distinctive:

    API keys: Verify env vars accessible to server process

    line 123
  7. Distinctive:

    Token refresh: Check token storage location and permissions

    line 124
  8. Distinctive:

    Lazy loading: Only enable servers needed for current task

    line 198

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MCP Server Orchestrator

Manage MCP server infrastructure for AI-powered development workflows.

MCP Architecture Overview

┌─────────────────┐     ┌──────────────────┐     ┌─────────────────┐
│   MCP Client    │────▶│   MCP Server     │────▶│  External APIs  │
│ (Claude, etc.)  │◀────│  (Tool Provider) │◀────│  (Services)     │
└─────────────────┘     └──────────────────┘     └─────────────────┘
         │                       │
         └───── JSON-RPC ────────┘

Key concepts:

  • Server: Provides tools, resources, and prompts via MCP protocol
  • Client: Consumes server capabilities (Claude Desktop, Claude Code, etc.)
  • Transport: Communication layer (stdio, SSE, WebSocket)

Configuration Locations

ClientConfig FilePlatform
Claude Desktopclaude_desktop_config.jsonmacOS: ~/Library/Application Support/Claude/
Windows: %APPDATA%\Claude\
Claude Codesettings.json or MCP configProject-level or user settings
Clinecline_mcp_settings.jsonVS Code extension settings

Server Configuration Schema

{
  "mcpServers": {
    "server-name": {
      "command": "executable",
      "args": ["arg1", "arg2"],
      "env": {
        "API_KEY": "value"
      },
      "disabled": false
    }
  }
}

Common Server Types

Python Server (uvx):

{
  "my-python-server": {
    "command": "uvx",
    "args": ["--from", "package-name", "server-command"]
  }
}

Node Server (npx):

{
  "my-node-server": {
    "command": "npx",
    "args": ["-y", "@scope/package-name"]
  }
}

Local Development Server:

{
  "dev-server": {
    "command": "python",
    "args": ["-m", "my_server"],
    "env": {
      "DEBUG": "true"
    }
  }
}

Troubleshooting Workflow

Connection Issues

  1. Verify server starts independently:
    # Test Python server
    python -m my_server
    
    # Test Node server
    npx -y @scope/package-name
  2. Check logs:
    • Claude Desktop: ~/Library/Logs/Claude/mcp*.log
    • Look for JSON-RPC errors, connection timeouts
  3. Validate JSON config:
    python -c "import json; json.load(open('config.json'))"
  4. Common fixes:
    • Use absolute paths for commands
    • Ensure dependencies installed in correct environment
    • Check API keys/env vars are set
    • Restart client after config changes

Authentication Issues

  1. OAuth flows: Ensure redirect URIs configured correctly
  2. API keys: Verify env vars accessible to server process
  3. Token refresh: Check token storage location and permissions

Building Custom Servers

Python Server (FastMCP)

from fastmcp import FastMCP

mcp = FastMCP("my-server")

@mcp.tool()
def my_tool(param: str) -> str:
    """Tool description for the AI."""
    return f"Result: {param}"

@mcp.resource("resource://my-data")
def get_data() -> str:
    """Provide data as a resource."""
    return "Resource content"

if __name__ == "__main__":
    mcp.run()

Node Server (MCP SDK)

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";

const server = new Server({ name: "my-server", version: "1.0.0" }, {
  capabilities: { tools: {} }
});

server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [{
    name: "my_tool",
    description: "Tool description",
    inputSchema: { type: "object", properties: { param: { type: "string" } } }
  }]
}));

const transport = new StdioServerTransport();
await server.connect(transport);

Multi-Server Orchestration

Modular Architecture

Organize servers by domain:

{
  "mcpServers": {
    "filesystem": { "command": "...", "args": ["--allowed-dirs", "/projects"] },
    "database": { "command": "...", "env": { "DB_URL": "..." } },
    "api-integrations": { "command": "...", "env": { "API_KEYS": "..." } },
    "custom-tools": { "command": "python", "args": ["-m", "my_tools"] }
  }
}

Server Selection Strategy

Think of servers as modules in a synthesizer—patch them together based on workflow needs:

  • Development workflow: filesystem + git + code-analysis servers
  • Research workflow: web-search + document + note-taking servers
  • Data workflow: database + visualization + export servers

Performance Optimization

  • Lazy loading: Only enable servers needed for current task
  • Caching: Implement response caching for expensive operations
  • Timeout tuning: Adjust timeouts for slow external APIs
  • Connection pooling: Reuse connections in database servers

References

  • references/server-templates.md - Boilerplate for common server types
  • references/debugging-guide.md - Detailed troubleshooting procedures

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.