Slide 1

Slide 1 text

MCP, A2A, and AG-UI: Architecting Agent Intelligence Beyond Chat MLCon Berlin 2026 Max Marschall @MaxOSchulte Consultant / Architect @ Thinktecture

Slide 2

Slide 2 text

Model Context Protocol (MCP) In this Session The Agent–User Interaction (AG-UI) Protocol Agent-to-Agent Protocol (A2A)

Slide 3

Slide 3 text

The Model Context Protocol Think of MCP like a USB-C port for AI applications. [...] MCP provides a standardized way to connect AI applications to external systems. https://modelcontextprotocol.io/docs/getting-started/intro

Slide 4

Slide 4 text

HOST • LLM applications (Claude Desktop, IDEs) that initiate connections The Model Context Protocol CLIENT • 1:1 connection handler inside host, manages protocol communication SERVER • Provides context, tools, and capabilities to clients • Servers can't talk to each other

Slide 5

Slide 5 text

• Do not repeat your self • (KISS) • No standardization across implementations • Fragile when systems change Why should we care?

Slide 6

Slide 6 text

M x N ➡ M+N App 1 App 3 App 2 Database Documentation FAQ App 1 App 3 App 2 Database Documentation FAQ MCP M AI Applications N Tools, Ressources, Prompts ➡

Slide 7

Slide 7 text

• Build once, connect everywhere • Open-source and vendor-neutral • Based on proven patterns The MCP Idea 💡

Slide 8

Slide 8 text

The Model Context Proctocol

Slide 9

Slide 9 text

The Model Context Proctocol

Slide 10

Slide 10 text

The Model Context Proctocol

Slide 11

Slide 11 text

• Tools • Ressources • Prompts • Elicitation • Sampling • Server Composition • Progress Report Features More to come (e.g. MCP-UI)

Slide 12

Slide 12 text

... is the ability to connect a large language model (LLM) to reliable sources of information to ensure accurate and relevant results. • Improves relevance • Prevents hallucinations • Connects to real-world • Alternatives: Retrievel-Augmented Generation (RAG) & Fine-Tuning Grounding

Slide 13

Slide 13 text

... is to provide an automated way to add concise contextual information based on the changing requirements. Examples: • Web search / web access • Documentation access Context-Hydration

Slide 14

Slide 14 text

https://modelcontextprotocol.info/docs/clients/ Feature Support Missing features Clients from MCP proposer

Slide 15

Slide 15 text

• Limit your tools • Scope your tools • Tool compatibility • Feature compatibility Best Practices

Slide 16

Slide 16 text

• Different tools by authentication / authorization • Update tool lists depending on preceding queries • Agent Skill enable / disable MCPs • Tool change (command) Context Size Solution: Dynamic Tools

Slide 17

Slide 17 text

• Async-Tools • MCP-UI • Code execution • (Antrophic / Claude) Agent Skills + MCP = RAG-MCP. Not “more tools,” but a better relationship to tools Outlook

Slide 18

Slide 18 text

Agent-to-Agent Protocol Model Context Protocol (MCP) The Agent–User Interaction (AG-UI) Protocol Agent-to-Agent Protocol (A2A)

Slide 19

Slide 19 text

A2A An open standard enabling secure, interoperable communication between autonomous AI agents across diverse frameworks and platforms

Slide 20

Slide 20 text

• Invented by Google • Gifted to Linux Foundation • "Build with ADK (or any framework), equip with MCP (or any tool), and communicate with A2A, to remote agents, local agents, and humans." A2A Origins https://a2a-protocol.org/latest/

Slide 21

Slide 21 text

Partners https://developers.googleblog.com/en/a2a-a-new-era-of-agent-interoperability/

Slide 22

Slide 22 text

A2A Layer https://a2a-protocol.org/latest/

Slide 23

Slide 23 text

• A2A treats agents as standard enterprise applications, relying on established web security practices. Identity handled at HTTP transport layer. • Production deployments MUST use HTTPS • OAuth 2.0, OpenID Connect, API Keys, Manual TLS • OpenAPI Aligned: Follows OpenAPI Security Scheme speci fi cation • Server Responsibility: Authenticate every request • In-Task Auth: Use auth-required state for secondary credentials during task execution Security & Authentication

Slide 24

Slide 24 text

Goals Interoperability Bridge communication gaps between disparate agentic systems Discovery Enterprise-grade secure communication patterns Security Enterprise-grade secure communication patterns Collaboration Enable agents to delegate tasks, exchange context, work together

Slide 25

Slide 25 text

• Simple: Reuses HTTP, JSON-RPC 2.0, Server-Sent Events • Enterprise Ready: Auth, security, privacy, tracing, monitoring built-in • Async First: Designed for long-running tasks, human-in-the-loop • Modality Agnostic: Text, audio/video, structured data, embedded UI • Opaque Execution: No need to share internal thoughts, plans, or tools Guiding Principles

Slide 26

Slide 26 text

Client • Discover agents via Agent Cards • Authenticate requests • Send messages / create tasks • Handle streaming responses • Expose webhook for noti fi cations Client Agent vs Remote Agent Remote • Publish Agent Card /.well - known/agent - card.json • Authenticate incoming requests • Process tasks & generate artifacts • Stream responses via SSE • Send push noti fi cations

Slide 27

Slide 27 text

Core Concepts 1. Agent Card Metadata, self-describing manifest 2. Task Stateful work unit, lifecycle, contains status, history and artifacts 3. Message Message ("user" / "agent"), 1-X parts, conversation history 4. Part Smallest Unit, TextPart, FilePart, DataPart 5. Artifact Generated result output, composed parts 6. Context Identi fi er to group tasks, maintains state across interactions

Slide 28

Slide 28 text

Agent Card https://a2a-protocol.org/latest/tutorials/python/3-agent-skills-and-card/#agent-skills

Slide 29

Slide 29 text

• id Unique task identi fi er • contextId Grouping identi fi er • status Current TaskStatus submitted ➡ working ➡ completed / failed / canceled / rejected • Interrupts: input-required & auth-required • history Message array • artifacts Generated outputs • metadata Extension data Task

Slide 30

Slide 30 text

• tasks/send - Submit task message to agent • tasks/get - Retrieve task status • tasks/cancel - Cancel running task • tasks/sendSubscribe - Send and subscribe to updates Core Protocol Methods

Slide 31

Slide 31 text

{ "jsonrpc": "2.0", "id": "req-001", "method": "message.send", "params": { "message": { "role": "user", "parts": [ { "kind": "text", "text": "Generate an image of a sailboat on the ocean." } ] "messageId": "msg-user-001" } } } Task Followup { "jsonrpc": "2.0", "id": "req-001", "result": { "id": "task-boat-gen-123", "contextId": "ctx-conversation-abc", "status": { "state": "completed" }, "artifacts": [ { "artifactId": "artifact-boat-v1-xyz", "name": "sailboat_image.png", "description": "A generated image of a sailboat on the ocean.", "parts": [ { "kind": "file", "file": { "name": "sailboat_image.png", "mimeType": "image/png", "bytes": "base64_encoded_png_data_of_a_sailboat" } } ] } ], { "jsonrpc": "2.0", "id": "req-002", "method": "message.send", "params": { "message": { "role": "user", "messageId": "msg-user-002", "contextId": "ctx-conversation-abc", "referenceTaskIds": [ "task-boat-gen-123" ], "parts": [ { "kind": "text", "text": "Please modify the sailboat to be red." } ] } } } { "jsonrpc": "2.0", "id": "req-002", "result": { "id": "task-boat-color-456", "contextId": "ctx-conversation-abc", "status": { "state": "completed" }, "artifacts": [ { "artifactId": "artifact-boat-v2-red-pqr", "name": "sailboat_image.png", "description": "A generated image of a red sailboat on the oce "parts": [ { "kind": "file", "file": { "name": "sailboat_image.png", "mimeType": "image/png", "bytes": "base64_encoded_png_data_of_a_RED_sailboat" } } ] } ], "kind": "task" } }

Slide 32

Slide 32 text

A2A Agentic Applications https://a2a-protocol.org/latest/topics/a2a-and-mcp/#a2a-mcp-complementary-protocols-for-agentic-systems

Slide 33

Slide 33 text

Key Takeaways • Open standard for agent interoperability • Built on existing web standards • Enterprise-grade security built-in • Complements MCP for complete stack Agent-to-Agent Protocol (A2A) Of fi cial Resources a2a-protocol.org github.com/a2aproject/A2A https://github.com/a2aproject/a2a- inspector https://www.a2aprotocol.net/docs/ speci fi cation

Slide 34

Slide 34 text

The Full Protocol Stack Model Context Protocol (MCP) The Agent–User Interaction (AG-UI) Protocol Agent-to-Agent Protocol (A2A)

Slide 35

Slide 35 text

The Agent–User Interaction (AG-UI) Protocol AG-UI is an open, lightweight, event-based protocol that standardizes how AI agents connect to user-facing applications. https://docs.ag-ui.com/introduction

Slide 36

Slide 36 text

• MCP connects agents to tools & context • A2A enables agent-to-agent communication • No standard for agent-to-user interaction • Each framework invented custom protocols Missing Link

Slide 37

Slide 37 text

Traditional APIs Don't Work for Agents Agentic applications break the request/response model. Client makes request → server returns data → interaction ends. Agents don't work this way.

Slide 38

Slide 38 text

• Custom WebSocket formats per agent • Ad-hoc JSON parsing and text hacks • Reinvent adapters for each framework • Complex state synchronization WITHOUT AG-UI (PAIN POINTS)

Slide 39

Slide 39 text

Standardisations Without AG-UI With AG-UI Framework fragmentation Every UI must code adapters for LangGraph, CrewAI, Mastra, etc. (M×N problem) Single protocol; any framework works with any UI (M+N) State synchronization Manual polling or complex WebSocket logic JSON Patch state deltas, standardized sync Human-in-the-loop Custom approval workflows per app Standardized HITL events built into protocol Debugging Server-side logs only; UI-agent communication is a black box Full client-side event tracing, transparent execution Generative UI No standard for agents to specify UI components Built-in component rendering events

Slide 40

Slide 40 text

Ecosystem Pydantic AI

Slide 41

Slide 41 text

AG-UI Server https://ai.pydantic.dev/ui/ag-ui/#examples https://docs.copilotkit.ai/pydantic-ai/quickstart/pydantic-ai?path=exiting-agent • Major framework support • "Simple" one-line expose • Fine grained control possible

Slide 42

Slide 42 text

CopilotKit Framework • Manages Communication • Helps with state • Human in the loop (HITL) • Use ready react components AG-UI Client https://docs.copilotkit.ai/pydantic-ai/quickstart/pydantic-ai?path=exiting-agent

Slide 43

Slide 43 text

Server Communication Events Category Description Lifecycle Events Monitor the progression of agent runs Text Message Events Handle streaming textual content Tool Call Events Manage tool executions by agents State Management Events Synchronize state between agents and UI Activity Events Represent ongoing activity progress Special Events Support custom functionality Draft Events Proposed events under development https://docs.ag-ui.com/concepts/events ➡ ∑16 standard events

Slide 44

Slide 44 text

• SSE (Server-Sent Events) - Default, text-based, HTTP • Binary Protocol - High performance, custom serialization • WebSocket - Bidirectional, low latency • HTTP Polling - Fallback for restrictive networks Communication Transport

Slide 45

Slide 45 text

1. TEXT_MESSAGE_START { messageId, role: "assistant" } 2. TEXT_MESSAGE_CONTENT (×N) { messageId, delta: "Hello" } { messageId, delta: " world" } 3. TEXT_MESSAGE_END { messageId } Text / Tool Streaming Event Pattern 1. TOOL_CALL_START { toolCallId: "call_123", toolName: "search_database" } 2. TOOL_CALL_ARGS { toolCallId: "call_123", args: { query: "users" } } 3. TOOL_CALL_END { toolCallId: "call_123", result: [...] }

Slide 46

Slide 46 text

• Bi-directional state management • STATE_SNAPSHOT Complete state object for initialization or refresh • STATE_DELTA JSON Patch (RFC 6902) for ef fi cient incremental updates State

Slide 47

Slide 47 text

Static • Agent speci fi es exact component and props to render • E.g. Component is mapped to a speci fi c event in the UI UI Capbilities Declarative • Agent describes intent, framework selects components • Agent knows about components and props

Slide 48

Slide 48 text

• Simple request/response with no streaming • Batch processing without user interaction • Static content delivery • Ultra-low latency requirements (<10ms) No-Usecase

Slide 49

Slide 49 text

• Real-time Streaming Complexity • Tool Orchestration • Agents call functions, run code, hit APIs — UI must show progress and handle approval • Concurrency & Cancellation • Uni fi ed Interface: LangChain, CrewAI, Mastra all speak different dialects Solved Problems

Slide 50

Slide 50 text

• Keep Humans in the loop • Visual Feedback • Actual questions and commits • Increase acceptance through observability • Loading placeholder effect AG-UI

Slide 51

Slide 51 text

• Documentation: docs.ag-ui.com • GitHub: github.com/ag-ui-protocol/ag-ui • AG-UI Dojo: https://dojo.ag-ui.com/pydantic-ai/ AG-UI Ressources

Slide 52

Slide 52 text

It's a wrap! • Slides: https://thinktecture.com/max-marschall • Contact: @MaxOSchulte, [email protected]