How do I install the MCP Spec skill?

Run npx skillmds@latest add xiaolai/mcp-spec in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is MCP Spec safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with MCP Spec?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is MCP Spec free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published MCP Spec?

xiaolai (@xiaolai) published this skill. Their other Agent Skills are listed on their SkillMD profile.

MCP Spec by xiaolai — Agent Skill

Name: MCP Spec
Author: xiaolai

MCP Spec

Router skill for the Model Context Protocol (MCP) open spec, covering the protocol itself (JSON-RPC framing, capabilities, lifecycle), the client and server roles, the transport layers (stdio / streamable HTTP / SSE), and the core primitives (tools, resources, prompts, sampling, roots, completion). Use when the user asks about: writing an MCP server (in any language), writing an MCP client, the JSON-RPC protocol framing, the initialize handshake, capability negotiation, the stdio transport, the streamable HTTP transport, the SSE transport, defining a tool / resource / prompt, sampling, roots, or completion. Includes the TypeScript SDK (`@modelcontextprotocol/sdk`) and Python SDK (`mcp`). Skip: Anthropic's hosted MCP connector (use anthropic-platform-features), the user-facing Claude Connectors directory (use claude-connectors), Claude Code's `.mcp.json` config (use claude-code → SKILL-mcp).

xiaolai baa1c1f 24 files · 188.8 KB Updated Sep 17, 2026

File contents

Model Context Protocol (MCP) — Router

Field	Value
Source docs	modelcontextprotocol.io
Spec repo	modelcontextprotocol/modelcontextprotocol
TypeScript SDK	`@modelcontextprotocol/sdk`
Python SDK	`mcp`

This skill is auto-updated daily. A pipeline reads the upstream docs and rewrites the per-surface files below. Section structure is stable; content drifts to track upstream.

Dispatch table

Surface file	Read when the user asks about…
`SKILL-protocol.md`	JSON-RPC framing, the `initialize` handshake, capability negotiation, lifecycle, error codes, versioning
`SKILL-clients.md`	implementing an MCP client, calling tools, reading resources, the client-side of sampling
`SKILL-servers.md`	implementing an MCP server in TypeScript or Python, server-side capability advertisement, server lifecycle
`SKILL-transport.md`	stdio transport, streamable HTTP transport, SSE transport (legacy), choosing a transport
`SKILL-tools-resources-prompts.md`	defining tools (input schema, output schema), defining resources (URI templates, MIME types), defining prompts, sampling, roots, completion

This skill is auto-updated daily by a maintainer-run pipeline. File issues at xiaolai/anthropic-docs.

---name: mcp-specdescription: Router skill for the Model Context Protocol (MCP) open spec, covering the protocol itself (JSON-RPC framing, capabilities, lifecycle), the client and server roles, the transport layers (stdio / streamable HTTP / SSE), and the core primitives (tools, resources, prompts, sampling, roots, completion). Use when the user asks about: writing an MCP server (in any language), writing an MCP client, the JSON-RPC protocol framing, the initialize handshake, capability negotiation, the stdio transport, the streamable HTTP transport, the SSE transport, defining a tool / resource / prompt, sampling, roots, or completion. Includes the TypeScript SDK (`@modelcontextprotocol/sdk`) and Python SDK (`mcp`). Skip: Anthropic's hosted MCP connector (use anthropic-platform-features), the user-facing Claude Connectors directory (use claude-connectors), Claude Code's `.mcp.json` config (use claude-code → SKILL-mcp).---# Model Context Protocol (MCP) — Router| Field | Value ||---|---|| **Source docs** | [modelcontextprotocol.io](https://modelcontextprotocol.io) || **Spec repo** | [modelcontextprotocol/modelcontextprotocol](https://github.com/modelcontextprotocol/modelcontextprotocol) || **TypeScript SDK** | [`@modelcontextprotocol/sdk`](https://www.npmjs.com/package/@modelcontextprotocol/sdk) || **Python SDK** | [`mcp`](https://pypi.org/project/mcp/) |> **This skill is auto-updated daily.** A pipeline reads the upstream> docs and rewrites the per-surface files below. Section structure is> stable; content drifts to track upstream.## Dispatch table| Surface file | Read when the user asks about… ||---|---|| [`SKILL-protocol.md`](SKILL-protocol.md) | JSON-RPC framing, the `initialize` handshake, capability negotiation, lifecycle, error codes, versioning || [`SKILL-clients.md`](SKILL-clients.md) | implementing an MCP client, calling tools, reading resources, the client-side of sampling || [`SKILL-servers.md`](SKILL-servers.md) | implementing an MCP server in TypeScript or Python, server-side capability advertisement, server lifecycle || [`SKILL-transport.md`](SKILL-transport.md) | stdio transport, streamable HTTP transport, SSE transport (legacy), choosing a transport || [`SKILL-tools-resources-prompts.md`](SKILL-tools-resources-prompts.md) | defining tools (input schema, output schema), defining resources (URI templates, MIME types), defining prompts, sampling, roots, completion |---*This skill is auto-updated daily by a maintainer-run pipeline. Fileissues at [xiaolai/anthropic-docs](https://github.com/xiaolai/anthropic-docs).*

Frequently asked questions