Agentic UI with AngularArchitecting Agentic AI with Open Standards
New eBook
Build agent-driven Angular UIs with CopilotKit and open standards, decoupled from any backend or model.
Integrate agents into Angular using open standards such as AG-UI, A2UI, and MCP Apps, adaptable user interfaces, human-in-the-loop patterns, and an architecture that remains independent of your backend stack.
Instant download Free updates Vendor-independent
You will get the following files:
- EPUB
Shipping a chat demo is easy — building a real agentic application isn’t
Wiring a model into a chat window takes an afternoon. The hard part is everything after: agents that reliably pursue your users’ goals, users who stay in control, and a frontend that doesn’t depend on this quarter’s vendor. That’s the gap this book closes.
Based on experience from 1,200+ workshops — trusted by teams at leading enterprises:
Details
Living eBook
Agentic AI is evolving rapidly — standards, tools, and patterns improve every few months.
This book is continuously updated to reflect the current state of AG-UI, A2UI, MCP, and CopilotKit.
All updates are included after purchase at no additional cost.
Guiding Principles
- ✅ Standards-First – open protocols like AG-UI, A2UI, and MCP Apps instead of vendor lock-in
- ✅ Architecture-First – keep your agentic frontend maintainable and integrate human-in-the-loop patterns
- ✅ Practice-First – every concept is backed by concrete implementation
What will I get?
- ☑️ Agent integration in Angular with CopilotKit
- ☑️ Client-side and server-side tool calling
- ☑️ Generative and dynamic UIs driven by agent responses
- ☑️ Patterns for production-grade systems
Target Audiences
- 1️⃣ Angular developers who want to integrate agents into their applications without coupling them to a specific model or backend
- 2️⃣ Architects and tech leads who need proven patterns for controllable, scalable agentic UIs
Prerequisites
Basic experience with Angular (components, services, templates) is needed.
No prior AI or machine learning knowledge is required — the book covers how agents work and the architecture of agentic UIs from the ground up.
Key Facts
- ✓ Regular updates included
- ✓ 240 pages
- ✓ Published: 2026
- ✓ Formats: PDF & EPUB
What’s Inside?
- ☑️ How agents work
- ☑️ Architecture of agentic UIs
- ☑️ Server-side and client-side tool calling
- ☑️ Structured outputs for dynamic components
- ☑️ Forms of agentic UIs
- ☑️ Testing agentic frontends
- ☑️ Connecting agents via AG-UI and CopilotKit
- ☑️ Dynamic and generative UI with A2UI
- ☑️ Custom catalogs with A2UI
- ☑️ MCP Apps for visual tool integration
- ☑️ Human-in-the-loop patterns
- ☑️ State for agents with AG-UI
- ☑️ Approval vs. co-planning and co-creation
- ☑️ Interrupt, undo, and action cards
- ☑️ Workflows and multi-agent architectures
- ☑️ Security and safety with guardrails
- ☑️ Multimodality: voice and images
What people are saying
“The needed step for the Agentic UI future.”
“The approach of building all concepts and features around a single, relatable business scenario makes the learning path easy to follow.”
“The book provides a highly illustrative and comprehensive overview of the topic of Agentic UI using practical Angular examples, which is why I can highly recommend it.”
“This well-structured book is recommended for developers who want to integrate AI support into their projects—suitable for both beginners and advanced users in this field.”
“Manfred’s concise, friendly presentation style, and clear code examples, make learning any concept accessible and enjoyable.”
“The book is a great collection of best practices for implementing LLM-based solutions in Angular. It gives a huge confidence boost when it comes to bringing AI integrations into production code.”
“Great resource if you want to explore how to use agentic UI in Angular, since most of the resources are targeting React.”
FAQs
How does VAT and invoicing work?
The displayed price is the final price.
VAT is handled automatically according to applicable regulations (based on your location and VAT status).
After purchase, you can enter your billing details (e.g., company name, address, VAT ID) and generate a VAT-compliant invoice.
Do I need prior AI or machine learning knowledge?
No.
The book explains how agents work, the architecture of agentic UIs, and tool calling from the ground up.
From there, the book moves step by step towards integrating agents into Angular and applying production-ready patterns.
Which Angular knowledge is expected?
Basic experience with Angular (components, services, templates) is sufficient.
The focus of the book is on integrating agents and designing agentic UIs — not on Angular fundamentals.
Am I locked into a specific model, backend, or AI vendor?
No — quite the opposite.
The book builds on open standards such as AG-UI, A2UI, and MCP Apps, so your frontend stays decoupled from any specific model or backend stack.
You can swap models or backends anytime without reworking your Angular application.
Why not just use free resources?
There are many free resources about AI and agents.
However, they usually focus on individual APIs or demos — not on how to structure real-world agentic frontends.
This book is based on experience from 1,200+ workshops with enterprise teams and focuses on architecture, controllability, and vendor independence.
It helps you connect the dots and avoid costly mistakes when bringing AI into real-world systems.
Do you offer workshops or services based on the book?
Yes.
We offer workshops, consulting, and architecture reviews based on the concepts in this book — with a strong focus on agentic UIs and scalable Angular architectures.
If you’re interested in working with us, feel free to get in touch.
About the Author
Manfred Steyer is a Google Developer Expert and Angular architect working with teams at leading enterprises. Together with his team at Angular Architects, he has conducted over 1,200 workshops for companies worldwide and is known for his practical, architecture-first approach.
He is a speaker at international conferences, an author for O’Reilly, and helps teams apply proven architecture patterns to AI-powered frontends.