MCP Deep Dive: The USB-C for AI Tools

Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers, and your agent can plug into almost any system. Here's how it actually works.

55 min · Reviewed 2026

Why MCP exists

Before MCP, every agent-to-tool connection was custom. Claude had one way to call a database. GPT had another. Your LangChain glue was a third. Multiply that by 1,000 tools and you had a combinatorial nightmare. Anthropic introduced MCP in late 2024 as 'USB-C for AI tools' — one protocol, any client, any server. By 2026 it's backed by Anthropic, OpenAI, Google, and the GitHub registry has 1,200+ servers.

The architecture

Piece	Role
Host	The app the user interacts with (Claude Desktop, Claude Code, Cursor, OpenClaw).
Client	The host's MCP client instance, one per connected server.
Server	A separate process exposing tools/resources/prompts.
Transport	stdio (spawned subprocess) or HTTP/SSE (remote service).
Protocol	JSON-RPC 2.0. Typed message schemas.

What a server exposes

Tools — function-like actions (call with args, get a result).
Resources — readable data URIs (file://, db://, custom://).
Prompts — reusable prompt templates the host can surface.
Sampling (optional) — server can ask host's model for completions.
Roots — URI hints about where the server operates.

A real MCP config

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-filesystem",
        "/Users/you/projects"
      ]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_..."
      }
    },
    "supabase": {
      "command": "npx",
      "args": [
        "-y",
        "@supabase/mcp-server-supabase@latest",
        "--project-ref",
        "your-project-ref"
      ],
      "env": {
        "SUPABASE_ACCESS_TOKEN": "sbp_..."
      }
    },
    "notion": {
      "url": "https://mcp.notion.com/sse",
      "transport": "sse",
      "headers": {"Authorization": "Bearer ntn_..."}
    }
  }
}Claude Desktop / Claude Code MCP config. Four real servers in the April 2026 registry. Drop this into ~/.claude/claude_desktop_config.json and restart.

Writing your own MCP server

// A tiny MCP server in TypeScript (@modelcontextprotocol/sdk)
import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
  CallToolRequestSchema,
  ListToolsRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

const server = new Server(
  { name: "my-server", version: "0.1.0" },
  { capabilities: { tools: {} } }
);

server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [{
    name: "greet",
    description: "Say hello to a person.",
    inputSchema: {
      type: "object",
      properties: { name: { type: "string" } },
      required: ["name"],
    },
  }],
}));

server.setRequestHandler(CallToolRequestSchema, async (req) => {
  if (req.params.name === "greet") {
    const name = (req.params.arguments as any)?.name ?? "friend";
    return { content: [{ type: "text", text: `Hello, ${name}!` }] };
  }
  throw new Error("Unknown tool");
});

const transport = new StdioServerTransport();
await server.connect(transport);A complete, minimal MCP server. Registers one tool, speaks stdio. About 30 lines.

Servers worth knowing (April 2026)

modelcontextprotocol/server-filesystem — scoped file access.
modelcontextprotocol/server-github — issues, PRs, repos.
modelcontextprotocol/server-postgres — SQL over any Postgres.
Supabase, Neon, Vercel, Stripe — all ship official first-party MCPs.
Zapier MCP — proxies to 8,000+ Zapier-connected apps.
Notion, Linear, Slack, Google Drive, Calendar — official.
Chroma, Pinecone, Vectara — vector DBs for RAG.
Sentry, Datadog — observability from inside your agent.

Transports

Transport	Use case	Pros	Cons
stdio	Local-only, spawned by host.	Zero network; trivial auth.	One process per client.
SSE / HTTP	Remote or shared server.	Multi-tenant, scales.	Need auth, TLS, ops.
Streamable HTTP (2025+)	Bidirectional over HTTP/2.	Best of both; modern.	Newer, fewer libs.

MCP is the one thing to learn cold. Once you have it, every agent gets cheaper, every tool is reusable, and switching models becomes a config change.

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-agentic-mcp-deep-dive-creators

What is the core idea behind "MCP Deep Dive: The USB-C for AI Tools"?
1. Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers, and your agent can plug into almost any system. Here's how it actually works.
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
Which term best describes a foundational idea in "MCP Deep Dive: The USB-C for AI Tools"?
1. stdio transport
2. MCP
3. SSE transport
4. MCP server
A learner studying MCP Deep Dive: The USB-C for AI Tools would need to understand which concept?
1. MCP
2. SSE transport
3. stdio transport
4. MCP server
Which of these is directly relevant to MCP Deep Dive: The USB-C for AI Tools?
1. MCP
2. stdio transport
3. MCP server
4. SSE transport
Which of the following is a key point about MCP Deep Dive: The USB-C for AI Tools?
1. Tools — function-like actions (call with args, get a result).
2. Resources — readable data URIs (file://, db://, custom://).
3. Prompts — reusable prompt templates the host can surface.
4. Sampling (optional) — server can ask host's model for completions.
Which of these does NOT belong in a discussion of MCP Deep Dive: The USB-C for AI Tools?
1. Prompts — reusable prompt templates the host can surface.
2. internal
3. Resources — readable data URIs (file://, db://, custom://).
4. Tools — function-like actions (call with args, get a result).
Which statement is accurate regarding MCP Deep Dive: The USB-C for AI Tools?
1. modelcontextprotocol/server-github — issues, PRs, repos.
2. modelcontextprotocol/server-postgres — SQL over any Postgres.
3. modelcontextprotocol/server-filesystem — scoped file access.
4. Supabase, Neon, Vercel, Stripe — all ship official first-party MCPs.
Which of these does NOT belong in a discussion of MCP Deep Dive: The USB-C for AI Tools?
1. modelcontextprotocol/server-github — issues, PRs, repos.
2. modelcontextprotocol/server-postgres — SQL over any Postgres.
3. modelcontextprotocol/server-filesystem — scoped file access.
4. internal
What is the key insight about "MCP servers are code you run" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. Install MCP servers the way you'd install packages — from trusted publishers only.
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
What is the key insight about "The deepest integration is LangGraph's" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. As of 2026, LangGraph treats MCP tools as first-class graph nodes with full streaming.
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
What is the key warning about "Scope your agents tightly" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. Always define: goal, tools, permissions, and stop condition before executing.
4. Document parameter constraints, units, and examples
Which statement accurately describes an aspect of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. Document parameter constraints, units, and examples
4. Before MCP, every agent-to-tool connection was custom. Claude had one way to call a database. GPT had another.
What does working with MCP Deep Dive: The USB-C for AI Tools typically involve?
1. MCP is the one thing to learn cold. Once you have it, every agent gets cheaper, every tool is reusable, and switching models becomes a confi…
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
Which best describes the scope of "MCP Deep Dive: The USB-C for AI Tools"?
1. It is unrelated to agentic workflows
2. It focuses on Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers,
3. It applies only to the opposite beginner tier
4. It was deprecated in 2024 and no longer relevant
Which section heading best belongs in a lesson about MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. The architecture
4. Document parameter constraints, units, and examples

← Back to interactive lesson

Tendril · Creators · Agentic AI

MCP Deep Dive: The USB-C for AI Tools

Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers, and your agent can plug into almost any system. Here's how it actually works.

55 min · Reviewed 2026

Why MCP exists

The architecture

Piece	Role
Host	The app the user interacts with (Claude Desktop, Claude Code, Cursor, OpenClaw).
Client	The host's MCP client instance, one per connected server.
Server	A separate process exposing tools/resources/prompts.
Transport	stdio (spawned subprocess) or HTTP/SSE (remote service).
Protocol	JSON-RPC 2.0. Typed message schemas.

What a server exposes

Tools — function-like actions (call with args, get a result).
Resources — readable data URIs (file://, db://, custom://).
Prompts — reusable prompt templates the host can surface.
Sampling (optional) — server can ask host's model for completions.
Roots — URI hints about where the server operates.

A real MCP config

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-filesystem",
        "/Users/you/projects"
      ]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_..."
      }
    },
    "supabase": {
      "command": "npx",
      "args": [
        "-y",
        "@supabase/mcp-server-supabase@latest",
        "--project-ref",
        "your-project-ref"
      ],
      "env": {
        "SUPABASE_ACCESS_TOKEN": "sbp_..."
      }
    },
    "notion": {
      "url": "https://mcp.notion.com/sse",
      "transport": "sse",
      "headers": {"Authorization": "Bearer ntn_..."}
    }
  }
}Claude Desktop / Claude Code MCP config. Four real servers in the April 2026 registry. Drop this into ~/.claude/claude_desktop_config.json and restart.

Writing your own MCP server

// A tiny MCP server in TypeScript (@modelcontextprotocol/sdk)
import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
  CallToolRequestSchema,
  ListToolsRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

const server = new Server(
  { name: "my-server", version: "0.1.0" },
  { capabilities: { tools: {} } }
);

server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [{
    name: "greet",
    description: "Say hello to a person.",
    inputSchema: {
      type: "object",
      properties: { name: { type: "string" } },
      required: ["name"],
    },
  }],
}));

server.setRequestHandler(CallToolRequestSchema, async (req) => {
  if (req.params.name === "greet") {
    const name = (req.params.arguments as any)?.name ?? "friend";
    return { content: [{ type: "text", text: `Hello, ${name}!` }] };
  }
  throw new Error("Unknown tool");
});

const transport = new StdioServerTransport();
await server.connect(transport);A complete, minimal MCP server. Registers one tool, speaks stdio. About 30 lines.

Servers worth knowing (April 2026)

modelcontextprotocol/server-filesystem — scoped file access.
modelcontextprotocol/server-github — issues, PRs, repos.
modelcontextprotocol/server-postgres — SQL over any Postgres.
Supabase, Neon, Vercel, Stripe — all ship official first-party MCPs.
Zapier MCP — proxies to 8,000+ Zapier-connected apps.
Notion, Linear, Slack, Google Drive, Calendar — official.
Chroma, Pinecone, Vectara — vector DBs for RAG.
Sentry, Datadog — observability from inside your agent.

Transports

Transport	Use case	Pros	Cons
stdio	Local-only, spawned by host.	Zero network; trivial auth.	One process per client.
SSE / HTTP	Remote or shared server.	Multi-tenant, scales.	Need auth, TLS, ops.
Streamable HTTP (2025+)	Bidirectional over HTTP/2.	Best of both; modern.	Newer, fewer libs.

MCP is the one thing to learn cold. Once you have it, every agent gets cheaper, every tool is reusable, and switching models becomes a config change.

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-agentic-mcp-deep-dive-creators

What is the core idea behind "MCP Deep Dive: The USB-C for AI Tools"?
1. Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers, and your agent can plug into almost any system. Here's how it actually works.
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
Which term best describes a foundational idea in "MCP Deep Dive: The USB-C for AI Tools"?
1. stdio transport
2. MCP
3. SSE transport
4. MCP server
A learner studying MCP Deep Dive: The USB-C for AI Tools would need to understand which concept?
1. MCP
2. SSE transport
3. stdio transport
4. MCP server
Which of these is directly relevant to MCP Deep Dive: The USB-C for AI Tools?
1. MCP
2. stdio transport
3. MCP server
4. SSE transport
Which of the following is a key point about MCP Deep Dive: The USB-C for AI Tools?
1. Tools — function-like actions (call with args, get a result).
2. Resources — readable data URIs (file://, db://, custom://).
3. Prompts — reusable prompt templates the host can surface.
4. Sampling (optional) — server can ask host's model for completions.
Which of these does NOT belong in a discussion of MCP Deep Dive: The USB-C for AI Tools?
1. Prompts — reusable prompt templates the host can surface.
2. internal
3. Resources — readable data URIs (file://, db://, custom://).
4. Tools — function-like actions (call with args, get a result).
Which statement is accurate regarding MCP Deep Dive: The USB-C for AI Tools?
1. modelcontextprotocol/server-github — issues, PRs, repos.
2. modelcontextprotocol/server-postgres — SQL over any Postgres.
3. modelcontextprotocol/server-filesystem — scoped file access.
4. Supabase, Neon, Vercel, Stripe — all ship official first-party MCPs.
Which of these does NOT belong in a discussion of MCP Deep Dive: The USB-C for AI Tools?
1. modelcontextprotocol/server-github — issues, PRs, repos.
2. modelcontextprotocol/server-postgres — SQL over any Postgres.
3. modelcontextprotocol/server-filesystem — scoped file access.
4. internal
What is the key insight about "MCP servers are code you run" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. Install MCP servers the way you'd install packages — from trusted publishers only.
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
What is the key insight about "The deepest integration is LangGraph's" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. As of 2026, LangGraph treats MCP tools as first-class graph nodes with full streaming.
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
What is the key warning about "Scope your agents tightly" in the context of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. Always define: goal, tools, permissions, and stop condition before executing.
4. Document parameter constraints, units, and examples
Which statement accurately describes an aspect of MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. Document parameter constraints, units, and examples
4. Before MCP, every agent-to-tool connection was custom. Claude had one way to call a database. GPT had another.
What does working with MCP Deep Dive: The USB-C for AI Tools typically involve?
1. MCP is the one thing to learn cold. Once you have it, every agent gets cheaper, every tool is reusable, and switching models becomes a confi…
2. internal
3. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
4. Document parameter constraints, units, and examples
Which best describes the scope of "MCP Deep Dive: The USB-C for AI Tools"?
1. It is unrelated to agentic workflows
2. It focuses on Model Context Protocol is the most important open standard in agents. One protocol, 1,200+ servers,
3. It applies only to the opposite beginner tier
4. It was deprecated in 2024 and no longer relevant
Which section heading best belongs in a lesson about MCP Deep Dive: The USB-C for AI Tools?
1. internal
2. Design per-task budgets for tool calls, tokens, and wall time so agents fail lou…
3. The architecture
4. Document parameter constraints, units, and examples

← Back to interactive lesson