Content
# A Survey of the Model Context Protocol (MCP): A Framework for Standardized, Agentic Interactions between LLMs and External Systems
---
## 📑 Table of Contents
- [A Survey of the Model Context Protocol (MCP): A Framework for Standardized, Agentic Interactions between LLMs and External Systems](#a-survey-of-the-model-context-protocol-mcp-a-framework-for-standardized-agentic-interactions-between-llms-and-external-systems)
- [📑 Table of Contents](#-table-of-contents)
- [Model Context Protocol (MCP): Empowering Agentic AI Interactions](#model-context-protocol-mcp-empowering-agentic-ai-interactions)
- [Comparative Analysis: MCP vs Traditional APIs](#comparative-analysis-mcp-vs-traditional-apis)
- [MCP Architecture Overview](#mcp-architecture-overview)
- [MCP Core Concepts](#mcp-core-concepts)
- [Resources](#resources)
- [Prompts](#prompts)
- [Tools](#tools)
- [Sampling](#sampling)
- [Roots](#roots)
- [Transport Layer](#transport-layer)
- [Agentic AI \& Composability](#agentic-ai--composability)
- [End-to-End MCP Workflow](#end-to-end-mcp-workflow)
- [🧠 AI Frameworks with MCP Integration](#-ai-frameworks-with-mcp-integration)
- [Example Clients – Feature Support Matrix](#example-clients--feature-support-matrix)
- [Reference \& Third-Party Servers](#reference--third-party-servers)
- [📦 MCP SDK Information](#-mcp-sdk-information)
---
## Model Context Protocol (MCP): Empowering Agentic AI Interactions
The **Model Context Protocol (MCP)** provides a structured, standardized way for **Large Language Models (LLMs)** to seamlessly interact with external tools, resources, and systems—much like how APIs and Language Server Protocols revolutionized application integration. MCP empowers the next generation of **agentic AI** by enabling autonomous, secure, and context-rich interactions.
---
## Comparative Analysis: MCP vs Traditional APIs
| Feature | Traditional APIs | Model Context Protocol (MCP) |
| ---------------------- | -------------------- | ---------------------------- |
| **Tool Usage** | Manual, bespoke code | Dynamic, standardized calls |
| **Prompt Interaction** | Basic text-based | Structured and context-aware |
| **Context Handling** | Limited | Integrated, built-in |
| **Discovery** | Manual | Dynamic and introspective |
| **Security** | Varies widely | Enforced mechanisms |
---
## MCP Architecture Overview
<div align="center">
<img src="assets/overview-model-context-protocol.svg" width="80%" alt="Overview of MCP Client-Server Architecture" />
<p><em>Fig. 1: MCP Client-Server Architecture.</em></p>
</div>
---
## MCP Core Concepts
### Resources
- **Structured External Data:** Exposes content such as text, audio, PDFs, system logs, and databases.
- **Types:** Text Resources (e.g., JSON, source code) and Binary Resources (e.g., PDFs, videos).
- **Discovery:** Via endpoints like `resources/list` and URI templates.
### Prompts
- **Reusable Templates:** For standardized LLM interactions.
- **Dynamic Context Injection:** Supports arguments and multi-step workflows.
- **Access Points:** Via `prompts/list` and `prompts/get`.
### Tools
<div align="center">
<img src="assets/tool.svg" width="75%" alt="MCP Tools" />
<p><em>Fig. 2: Tools provide active invocation using defined JSON schemas.</em></p>
</div>
- **Executable Capabilities:** Trigger actions and external system calls.
- **Definition:** Each tool is defined with a name, description, input/output schema, and validation.
- **Invocation:** Accessed via `tools/list` and invoked using `tools/call`.
### Sampling
<div align="center">
<img src="assets/sampling.svg" width="75%" alt="MCP Sampling Flow" />
<p><em>Fig. 3: Secure and contextual LLM completions via MCP sampling.</em></p>
</div>
- **Server-Initiated:** Sends messages to the LLM through the client.
- **Human-in-the-Loop:** Incorporates review/approval for secure execution.
- **Control Parameters:** Enables fine-tuning (temperature, token limits, etc.).
### Roots
<div align="center">
<img src="assets/root.svg" width="75%" alt="MCP Roots" />
<p><em>Fig. 4: Roots define operational boundaries using URIs.</em></p>
</div>
- **Logical Boundaries:** Define scopes (directories, API endpoints) for resource access.
- **Multi-Context Support:** Enables composable, dynamic agent workflows.
### Transport Layer
- **Real-Time Communication:** Utilizes secure HTTP/SSE channels.
- **Reliable Messaging:** Ensures structured, bidirectional interaction.
---
## Agentic AI & Composability
MCP is a catalyst for **agentic AI**, enabling autonomous agents to interact, collaborate, and chain tasks dynamically.
- **Dual Role Components:** MCP nodes act as both clients and servers.
- **Dynamic Agent Chaining:** Supports complex workflows with an orchestrator triggering specialized sub-agents.
- **Human Oversight:** Built-in review processes ensure security and reliability.
<div align="center">
<img src="assets/composabilty_agentic.png" width="80%" alt="Agentic AI & Composability" />
<p><em>Fig. X: Dynamic agent chaining enabled by MCP composability.</em></p>
</div>
<div align="center">
<img src="assets/orchestrator.png" width="80%" alt="Agentic AI & Composability" />
<p><em>Fig. y: Multi Agent Orchestrator.</em></p>
</div>
---
## End-to-End MCP Workflow
<div align="center">
<img src="assets/mcp-use-case.png" width="90%" alt="MCP Workflow Example" />
<p><em>Fig. 5: Customer Support Chatbot Workflow powered by MCP.</em></p>
</div>
**Workflow Highlights:**
1. **Capture & Send:** Client submits a structured prompt.
2. **Secure Transport:** Data is exchanged via the Transport Layer.
3. **Dynamic Invocation:** Server retrieves tools and resources as needed.
4. **Sampling & Review:** Server requests LLM completions with human oversight.
5. **Response Generation:** Outputs are returned in a clear, structured format.
---
## 🧠 AI Frameworks with MCP Integration
| Framework | Description | Integration Link |
| -------------- | ------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------- |
| **LangChain** | Lightweight wrapper for making Anthropic MCP tools compatible with LangChain and LangGraph. | [langchain-ai/langchain-mcp-adapters](https://github.com/langchain-ai/langchain-mcp-adapters) |
| **CrewAI** | Provides an MCP server to manage and trigger deployed CrewAI workflows. | [crewAIInc/enterprise-mcp-server](https://github.com/crewAIInc/enterprise-mcp-server) |
| **LlamaIndex** | Integration for connecting LlamaIndex tools with MCP tools using `llama-index-tools-mcp`. | [llama-index-tools-mcp](https://github.com/run-llama/llama_index/tree/main/llama-index-integrations/tools/llama-index-tools-mcp) |
---
## Example Clients – Feature Support Matrix
| Client | Resources | Prompts | Tools | Sampling | Roots | Notes |
| ------------------ | :-------: | :-----: | :---: | :------: | :---: | ------------------------------------------------------------------- |
| Claude Desktop App | ✅ | ✅ | ✅ | ❌ | ❌ | Full support for all MCP features |
| 5ire | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| BeeAI Framework | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools in agentic workflows. |
| Cline | ✅ | ❌ | ✅ | ❌ | ❌ | Supports tools and resources. |
| Continue | ✅ | ✅ | ✅ | ❌ | ❌ | Full support for all MCP features |
| Cursor | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| Emacs Mcp | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools in Emacs. |
| Firebase Genkit | ⚠️ | ✅ | ✅ | ❌ | ❌ | Supports resource list and lookup through tools. |
| GenAIScript | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| Goose | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| LibreChat | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools for Agents. |
| mcp-agent | ❌ | ❌ | ✅ | ⚠️ | ❌ | Supports tools, server connection management, and agent workflows. |
| oterm | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| Roo Code | ✅ | ❌ | ✅ | ❌ | ❌ | Supports tools and resources. |
| Sourcegraph Cody | ✅ | ❌ | ❌ | ❌ | ❌ | Supports resources through OpenCTX. |
| Superinterface | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools. |
| TheiaAI/TheiaIDE | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools for Agents in Theia AI and the AI-powered Theia IDE. |
| Windsurf Editor | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools with AI Flow for collaborative development. |
| Zed | ❌ | ✅ | ❌ | ❌ | ❌ | Prompts appear as slash commands. |
| SpinAI | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools for Typescript AI Agents. |
| OpenSumi | ❌ | ❌ | ✅ | ❌ | ❌ | Supports tools in OpenSumi. |
| Daydreams Agents | ✅ | ✅ | ✅ | ❌ | ❌ | Support for drop-in servers to Daydreams agents. |
---
## Reference & Third-Party Servers
| Server | Description | Link |
| -------------------- | --------------------------------------------------------------------------- | ----------------------------------------------------------------------------------- |
| **AWS KB Retrieval** | Retrieves data from the AWS Knowledge Base using Bedrock Agent Runtime. | [GitHub](https://github.com/rishikavikondala/mcp-server-aws) |
| **Google Drive** | Enables file access and search within Google Drive. | [GitHub](https://github.com/modelcontextprotocol/servers/tree/main/src/gdrive) |
| **Google Maps** | Provides location services, directions, and place details. | [GitHub](https://github.com/modelcontextprotocol/servers/tree/main/src/google-maps) |
| **Redis** | Interacts with Redis key-value stores for caching and data management. | [GitHub](https://github.com/modelcontextprotocol/servers/tree/main/src/redis) |
| **PostgreSQL** | Offers read-only database access with schema inspection. | [GitHub](https://github.com/modelcontextprotocol/servers/tree/main/src/postgres) |
| **Cloudflare** | Deploys, configures, and interrogates resources on the Cloudflare platform. | [GitHub](https://github.com/cloudflare/mcp-server-cloudflare) |
| **Stripe** | Integrates with Stripe API to manage payments, customers, and refunds. | [GitHub](https://github.com/atharvagupta2003/mcp-stripe) |
| **Neo4j** | Provides interaction with Neo4j Graph Database for graph-based operations. | [GitHub](https://github.com/neo4j-contrib/mcp-neo4j/) |
| **Apify** | Leverages pre-built cloud tools to extract data from websites and APIs. | [GitHub](https://github.com/apify/actors-mcp-server) |
<div align="center">
**🌐 Source Repositories**
🔗 [modelcontextprotocol.io/clients](https://modelcontextprotocol.io/clients)
🔗 [github.com/modelcontextprotocol/servers](https://github.com/modelcontextprotocol/servers)
🔗 [mcp.composio.dev](https://mcp.composio.dev)
🔗 [mcp.so](https://mcp.so/)
</div>
---
## 📦 MCP SDK Information
| SDK Name | Language | Description | Link |
| ---------------------- | ---------- | ------------------------------------------------------------------- | ---------------------------------------------------------------- |
| **TypeScript MCP SDK** | TypeScript | A comprehensive SDK to build MCP servers and clients in TypeScript. | [GitHub](https://github.com/modelcontextprotocol/typescript-sdk) |
| **Python MCP SDK** | Python | A robust SDK for implementing MCP servers and clients in Python. | [GitHub](https://github.com/modelcontextprotocol/python-sdk) |
| **Java MCP SDK** | Java | SDK for building and managing MCP infrastructure in Java. | [GitHub](https://github.com/modelcontextprotocol/java-sdk) |
| **Kotlin MCP SDK** | Kotlin | Kotlin-native SDK for MCP client-server communication. | [GitHub](https://github.com/modelcontextprotocol/kotlin-sdk) |
| **C# MCP SDK** | C# | SDK for .NET/C# developers to integrate MCP functionality. | [GitHub](https://github.com/modelcontextprotocol/csharp-sdk) |
