Create and manage unicode braille animations and spinners for CLIs and web apps. Use this skill when the user wants to add loading indicators, progress animations, or custom braille-based art to their terminal or browser-based application.

2026-07-213

code-scaffold-harness

upioneer/code-scaffold

Agent harness for headlessly deploying Code Scaffold assets.

2026-07-213

codebase-memory-mcp

upioneer/code-scaffold

High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds.

2026-07-213

cybersecurity-toolkit

upioneer/code-scaffold

Comprehensive cybersecurity arsenal integrating MITRE and NIST framework methodologies, NVIDIA SkillSpector vulnerability scanning, and deep SAST secret detection.

2026-07-213

name	MCP Generator
description	Assists with creating new custom Model Context Protocol (MCP) servers, clients, and apps using official SDKs and best practices.

MCP Generator Skill

When this skill is executed, you are tasked with scaffolding a robust, secure, and dependable Model Context Protocol (MCP) project. Follow these exact steps:

Step 1: Requirements Gathering

Use your ask_question tool to determine the project specifications from the user:

Question 1: "What type of MCP component are you building?" Options:

MCP Server (Exposes Tools, Resources, Prompts)
MCP Client (Connects to servers to consume capabilities)
Full App (Both Client and Server)

Question 2: "Which language/ecosystem do you prefer?" Options:

TypeScript / Node.js
Python
Java
Kotlin
C#
Go
PHP
Ruby
Rust
Swift

Question 3: "Which transport protocol should be configured by default?" Options:

stdio (Standard Input/Output)
SSE (Server-Sent Events over HTTP)

(Note: The ask_question tool automatically provides a text input box in the UI, so do not add an "Other" option. Users can use that box to provide custom requirements).

Step 2: Architecture & Best Practices Review

Before writing any code, silently review the reference materials located in this skill's directory under references/. This ensures you are aligning with the official @modelcontextprotocol/sdk (TS) or mcp (Python) documentation, security guidelines, and protocol versioning rules.

Step 3: Scaffold the Project

Initialize the project structure based on the user's choices.

For TypeScript: Initialize package.json and install @modelcontextprotocol/sdk.
For Python: Initialize pyproject.toml or requirements.txt and install mcp.
For Java: Scaffold a pom.xml or build.gradle and use the java-sdk.
For Kotlin: Base the scaffold on the create-kotlin-server pattern or the kotlin-sdk.
For C#: Run dotnet new and integrate the csharp-sdk.
For Go: Run go mod init and fetch the go-sdk.
For PHP, Ruby, Rust, or Swift: Initialize their respective package managers (composer, bundle, cargo, swift package init) and integrate their language-specific SDKs.

Step 4: Implement the Core Logic

Write the boilerplate code for the server or client:

Server: Implement the Server class from the SDK, configure the chosen transport, and scaffold example handlers for tools (ListToolsRequestSchema, CallToolRequestSchema) and resources (ListResourcesRequestSchema, ReadResourceRequestSchema).
Client: Implement the Client class, connect via the chosen transport, and demonstrate fetching tools/resources and calling a tool.

Security Best Practices to enforce:

Validate all incoming tool arguments rigidly using a validation library (like zod in TS or pydantic in Python).
Do not expose sensitive environment variables through MCP resources unless explicitly requested.
Ensure proper error handling and return appropriate JSON-RPC error codes (using McpError and ErrorCode).

Step 5: Finalize and Document

Create a README.md in the generated project directory explaining how to build, run, and test the MCP component (e.g., using the MCP Inspector via npx @modelcontextprotocol/inspector). Inform the user that the generation is complete and provide instructions on how to test it.