In partnership with

What is MCP?

The Model Context Protocol (MCP), developed by Anthropic, is an open standard that connects AI applications with external data sources and tools. AI Twitter has taken to referring to MCP as the USB port of of AI agents or HTTP. Think of it as a universal translator that allows AI agents to seamlessly interact with various systems—databases, APIs, business tools—without requiring custom integrations for each one.

The other day I came across this great insight about how MCP enables agents to access data and in doing so disrupts businesses that rely on the attention monetization around ads. Further, MCPs make agents an intermediary between brands and their users. This technology shift has many profound implications for how business will be conducted online in the not so distant future.

The Benefits of MCP

MCP offers several transformative advantages:

1. Enhanced AI capabilities: Agents gain access to real-time, specialized data, dramatically improving their ability to solve complex problems.

2. Reduced integration complexity: Developers can build once and connect to many systems, accelerating AI application development.

3. New user experiences: Users can interact with multiple systems through consistent AI interfaces rather than jumping between different apps.

4. Data portability: Information becomes more accessible across systems, reducing lock-in to specific platforms.

5. Democratized AI development: Smaller companies can create sophisticated AI applications by leveraging existing data sources through MCP.

What is MCP, Really?

At its core, MCP functions as a universal translator between AI applications and external data sources. Developed by Anthropic, this open standard allows AI agents to seamlessly access:

• Real-time data from specialized sources

• Business tools and enterprise systems

• APIs and databases without custom integrations

This creates a profound shift: rather than forcing users to navigate multiple interfaces, MCP enables AI agents to become the primary interface through which we interact with digital services.

The Disruptive Potential

MCP disrupts traditional attention-based business models by positioning AI agents as intermediaries between brands and users. Companies that have built moats around their data face significant challenges:

LinkedIn: ~$2.1B (20.4%) of revenue potentially threatened as MCP could bypass connection governance structures.

Workday: ~$1.2B (19.2%) at risk as MCP challenges their control over enterprise data and user experience.

Salesforce: ~$4.8B (15%) jeopardized as MCP threatens their position as gatekeeper of customer data.

Getting Started with MCP: A Quick Start Guide

What is MCP?

Model Context Protocol (MCP) is a standardized way for LLMs to use tools via a client-server architecture. In everyday terms, it lets AI models like Claude safely connect with your apps and data.

Why Should You Care?

With MCP, you can build a personalized AI assistant that can:

• Access and work with your data where it already lives

• Connect with the tools you use daily

• Run locally and securely on your computer

• Eliminate copying and pasting between apps

Quick Start Guide 5 Minutes or less

Step 1: Install Claude Desktop

Download the latest version from Anthropic's website.

Step 2: Install an MCP Server

MCP servers are usually installed via:

npm install -g @example/mcp-server

or using Docker:

docker pull example/mcp-server
docker run -p 8000:8000 example/mcp-server

Step 3: Configure Claude to Use Your MCP Server

1. Open Claude Desktop

2. Go to Settings → Developer → Edit Config

3. Add your server to the configuration:

{
  "mcpServers": [
    {
      "name": "My MCP Server",
      "url": "http://localhost:8000"
    }
  ]
}

4. Save and restart Claude Desktop

Step 4: Start Using Your Tools

Claude will now recognize when you're talking about content or tasks related to your connected tools and offer to use them.

Try asking something like: "Can you summarize my recent notes about project X?"
(if you've connected Notion or another note-taking app)

Popular MCP Integrations

• Google Drive: Search and read your documents

• Notion: Work with your notes and databases

• Slack: Get Claude to analyze your conversations

• GitHub: Analyze code repositories

• Figma: Reference your designs (view only)

First Principles for Designers in an MCP World

As designers and technologists, we need to rethink fundamental assumptions:

1. Design for delegation, not just direct manipulation

   • How might interfaces change when users primarily interact through an agent?

   • What control mechanisms do users need over agent behavior?

2. Rethink information architecture for agent accessibility

   • What metadata makes your design components more accessible to agents?

   • How do you organize information to support both human and agent access patterns?

3. Consider the entire agent-mediated journey

   • How does your product fit into broader workflows that may span multiple systems?

   • What contextual information should persist across these boundaries?

The Path Forward

MCP represents a fundamental shift in how we design and build digital experiences. The most successful businesses won't be those with the largest walled gardens, but those that position themselves as valuable nodes in an interconnected network of agent-accessible services.

For designers and technologists, this means embracing a new paradigm where our creations may be primarily experienced through the lens of an AI agent rather than direct human interaction. This isn't a limitation—it's an opportunity to focus on delivering genuine value rather than capturing attention.

The question isn't whether to adapt to this new reality, but how quickly we can reimagine our products and services for an agent-first world.

Try Artisan’s All-in-one Outbound Sales Platform & AI BDR

Ava automates your entire outbound demand generation so you can get leads delivered to your inbox on autopilot. She operates within the Artisan platform, which consolidates every tool you need for outbound:

• 300M+ High-Quality B2B Prospects, including E-Commerce and Local Business Leads

• Automated Lead Enrichment With 10+ Data Sources

• Full Email Deliverability Management

• Multi-Channel Outreach Across Email & LinkedIn

• Human-Level Personalization

Book a demo to see what Ava can do.

💬 Suggestion box

A newsletter exploring the principles that will help us design AI agent experiences and startups that amplify human creativity.

Subscribe to join a community of designers and developers shaping purposeful AI agents.

Until next time, keep innovating and stay curious!

Jason