High-Performance MCP Meteorological Data Access Utility

An enterprise-grade scaffolding project designed for constructing Model Context Protocol (MCP) service providers utilizing the TypeScript language.

🚀 Rapid Initialization

Clone the source repository structure
Execute dependency resolution: bash bun install

✨ Core Capabilities

Leveraging Bun runtime for accelerated testing cycles and development iteration speed
Integrated static analysis and code formatting powered by Biome
Synchronized artifact revision tracking via standard-version tooling
Architected for maximum code clarity and maintainability

📂 Project Organization Schema

mcp-starter/ ├── src/ │ ├── tools/ # Core MCP tool implementations reside here │ ├── utils/ # Common, shared utility functions │ ├── main.ts # Application bootstrap and server entry point │ └── types.ts # Canonical type definitions for data contracts ├── tests/ # Unit and integration test suites ├── biome.json # Configuration file for code style enforcement ├── tsconfig.json # TypeScript compiler settings └── package.json # Manifest detailing project dependencies

⚙️ Custom Tool Generation

This template includes a utility script facilitating the bootstrapping of novel MCP functionalities:

bash bun run scripts/create-tool.ts

This command automates the following steps: 1. Provisioning a dedicated directory under src/tools/<new-tool-designation> 2. Populating the directory with boilerplate files: - index.ts (The primary implementation logic) - schema.ts (Definition of the input parameter JSON Schema) - test.ts (A foundational test case) 3. Registering the newly created tool within the main tools index file.

Instantiation Example: bash bun run scripts/create-tool.ts precipitation_forecast

🛠️ Operational Commands

Execute Tests: bun test
Enforce Formatting: bun run format
Perform Linting Checks: bun run lint
Compile Artifacts: bun run build

To integrate this operational MCP mechanism with the Claude Desktop client:

Compile the project artifacts: bash bun run build
Update your Claude Desktop configuration manifest:

// Adjust arguments here if the server binary requires specific runtime inputs { "mcpServers": { "my-weather-service": { "command": "node", "args": ["/absolute/path/to/project/dist/main.js", "optional_startup_flag"] } } }

📜 Artifact Revision Control

This codebase adheres to standard-version for systematic, automated version advancement. Invoke bun run release to commit changes, tag the repository, and update version manifests.

Commit Message Convention

feat: Introduction of a new capability (triggers minor version bump)
fix: Resolution of a defect (triggers patch version bump)
BREAKING CHANGE: A backward-incompatible modification (triggers major version bump)

📦 Distribution via npm Registry

Authenticate with the npm registry: bash npm login
Generate the deployable build assets: bash bun run build
Publish the compiled package: bash npm publish

Crucially, execute bun run release to formalize versioning prior to any public registry submission.

Installing via Smithery

To achieve immediate, automated deployment of the Weather MCP Server Starter into your Claude Desktop environment via Smithery:

bash npx -y @smithery/cli install @GreatAuk/mcp-weather --client claude

Installing from npm (Post-Publication)

Modify your Claude Desktop runtime configuration as follows:

// Include arguments here only if necessary for the CLI execution context { "mcpServers": { "deployed-weather-tool": { "command": "npx", "args": ["-y", "your-published-package-name", "cli_argument_value"] } } }