mcp-server-starter
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-server-starterwhat's the weather in New York?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-server-starter
A production-grade starting point for a Model Context Protocol server — stdio and Streamable HTTP from one codebase, zod-validated tools, real error handling, hermetic client-level tests, and an MCPB manifest for one-click install in Claude Desktop.
Two worked example tools, each teaching a real pattern:
get_weather— wrap an external API, shape the result, fail usefully (runs with zero API keys)notes_add / notes_list / notes_get / notes_delete— own state, CRUD, input validation, safe destructive naming
Quick start
npm install
npm run smoke # build + hermetic client-level tests (3/3 as shipped)
npm start # stdio — Claude Desktop, Cursor, ...
npm start -- --http # Streamable HTTP on http://localhost:3001/mcpThe Streamable HTTP path is wire-verified (initialize → tools/list → DELETE round trip against the current SDK).
Related MCP server: Server MCP
What this repo doesn't include
This starter ships free from the full kit, which adds:
the Python twin starter (same patterns, one file)
the tool-design guide — naming, schemas, failure modes, and why LLM clients misuse badly-designed tools
the July 2026 spec-transition map — the stateless core explained, what breaks (nothing yet), and the two independent migrations everyone confuses
a spec-driven generator — describe tools in JSON, get a working, test-verified server out
packaging + deploy configs —
.mcpbone-click packaging, Dockerfile, systemd unit, Procfilea service-intake template if you want to sell builds yourself
→ The full kit: https://wisetechysolutions.github.io/mcp-server-starter/
License
MIT for this repository. The full kit is a separate product under its own license.
This server cannot be deployed
Maintenance
Related MCP Connectors
Persistent memory, hybrid search and a goal graph for AI agents, over stdio or remote HTTP.
ADHD-friendly tasks, notes & projects for LucidNest - 18 tools, scoped tokens, Streamable HTTP.
Build, validate, deploy — HTTP APIs, cron jobs, webhooks and MCP tools — from your AI client.
Minimal chat-first app with durable threads, actions, and the app-agent loop
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI models to access external services including weather data, file system operations, and SQLite database interactions through a standardized JSON-RPC interface. Features production-ready architecture with security, rate limiting, and comprehensive error handling.318 npmMIT
- FlicenseNot gradedqualityDmaintenanceProvides math and weather tools accessible via LangGraph agent using MCP protocol with stdio and streamable HTTP transports.1-
- AlicenseNot gradedqualityBmaintenanceExposes weather data from the free Open-Meteo API via decorator-driven MCP tools with Zod schemas, supporting stdio and HTTP transports.17 npmMIT
- AlicenseNot gradedqualityCmaintenanceExposes one SQLite-backed notes store over both stdio and an HTTPS/SSE endpoint, letting Claude Code, Claude Desktop, Claude web, and Claude mobile read and write the same shared state with full-text search. Provides write, read, search, and list tools so a note created on one surface is immediately visible from any other.11 npmMIT