Skip to main content

​
What is the Model Context Protocol (MCP)?

MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external systems. Using MCP, AI applications like Claude or ChatGPT can connect to data sources (e.g. local files, databases), tools (e.g. search engines, calculators) and workflows (e.g. specialized prompts), enabling them to access key information and perform tasks. Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect electronic devices, MCP provides a standardized way to connect AI applications to external systems.

​
Core Components

​
MCP Servers

MCP servers are programs that expose specific capabilities to AI applications through standardized protocol interfaces. Common examples include file system servers for document access, database servers for data queries, GitHub servers for code management, Slack servers for team communication, and calendar servers for scheduling.

​
Tools

Tools enable AI models to perform actions. Each tool defines a specific operation with typed inputs and outputs. The model requests tool execution based on context.

​
Resources

Resources provide structured access to information that the AI application can retrieve and provide to models as context.

​
Prompts

Prompts provide reusable templates. They allow MCP server authors to provide parameterized prompts for a domain, or showcase how to best use the MCP server.

​
Authentication Providers

Authentication Providers handle security for MCP server and API access. AgentPass supports:
  • OAuth 2.0: Industry-standard authorization framework (OAuth providers)
  • JWT: JSON Web Token authentication (JWKS endpoints)
  • Custom Headers: Custom authentication schemes (Token-based authentication such as API keys, etc.)
Provider Hierarchy:
  1. Server-level: Default authentication for all tools
  2. Tool-level: Override server authentication for specific tools
  3. User-level: Personal credentials and OAuth tokens

​
Workflow Architecture

​
Data Flow

Here’s how data flows through the AgentPass ecosystem:
1

AI Agent Request

An AI agent (Claude, Cursor, etc.) makes a request to execute a tool through the MCP protocol.
2

MCP Server Processing

The MCP server receives the request and identifies the target tool and required parameters.
3

Authentication

The server applies the appropriate authentication (OAuth, API key, etc.) based on the tool configuration.
4

API Call

The authenticated request is made to the target API endpoint with transformed parameters.
5

Response Processing

The API response is processed, formatted, and returned to the AI agent via MCP.
6

Analytics & Logging

All interactions are logged and analyzed for monitoring and optimization.

​
Common Use Cases

API Gateway

Use MCP servers as gateways to internal APIs, providing controlled AI agent access with authentication and monitoring.

Data Integration

Connect AI agents to databases, CRMs, and other data sources with appropriate security and formatting.

Workflow Automation

Enable AI agents to trigger business processes, update records, and coordinate complex workflows.

Development Tools

Integrate with development platforms, CI/CD systems, and code repositories for AI-assisted development.

​
Next Steps

Now that you understand the core concepts, explore these areas: Understanding these concepts will help you make the most of AgentPass features for MCP servers integration and management.
⌘I