正在加载视频...

视频加载失败

Unreal Model Context Protocol (MCP) provides an official, Epic-supported way to connect your favorite agentic LLM tools directly to Unreal Engine🤝 Built on the open MCP standard, it enables agents to take action in your projects using the capabilities exposed by each editor—so you can use compatible agentic coding...

226,748 次观看 • 3 天前 •via X (Twitter)

31 条评论

Unreal Engine 的头像
Unreal Engine3 天前

⚫Learn how to connect MCP in UE: ⚫Learn how to connect MCP in UEFN: ⚫Explore PCG and MCP workflows with the UE City Sample:

Ziwen 的头像
Ziwen3 天前

I love unreal.. Can we make a CLI too?

Cameron Poe 的头像
Cameron Poe3 天前

You guys have a lot of work to do to catch up to the rest of the industry with AI integration and world models. You’re falling behind rapidly.

bugyaluwang 的头像
bugyaluwang3 天前

Why is your MCP a single instance? This makes it hard to handle multiple agents

e/a sporz itzindagaem 的头像
e/a sporz itzindagaem3 天前

AI got so good it doesn’t need MCPs anymore.

Neura 的头像
Neura3 天前

You're behind

-del1an 的头像
-del1an3 天前

Probably the worst "MCP" implementation. LLMs spend so much time exploring the tools, search/find commands, in fact, it spends more time doing that instead of the actual task.

Anderson 的头像
Anderson3 天前

game devs who skip AI tooling will lose to those who use it

Anon 的头像
Anon3 天前

pointless cause ai doesnt have blueprint access with this trash mcp. update it and give real access

LaDoger 🟧 的头像
LaDoger 🟧3 天前

YES YES YES

Derek J. Spohn 的头像
Derek J. Spohn2 天前

Been using this and the videos always make it seem like the agent flies through building high quality at insane speeds. I can tell you this is not the case. It’s slow and it’s painful, even with bringing in premade assets.

Pils10 的头像
Pils103 天前

Used it. My AI agent just gave up half way trough and was like "there is no tool to do that" and was like "do it manually in the UI". He did the C++ part super well, but when it came to the MCP part, he couldn't create most components and had issues re-parenting etc.

Keen 的头像
Keen3 天前

the demo i want to see: change 50 actors, inspect the result, undo the whole batch. agent workflows need a good ctrl+z story.

Roger Dottin 的头像
Roger Dottin3 天前

Wow, this is EPIC. I mean UNREAL.

Tom 的头像
Tom3 天前

Its good to see Unreal moving in this direction. I think they can see which way the tide is going. Excited to see how they keep improving Unreal to be SI agent focused. Its the future.

Slopmaxxer 的头像
Slopmaxxer2 天前

Let’s at least not recycle the demo from 6 months ago

Lex Luther 的头像
Lex Luther1 天前

Isn’t this just the experimental mcp you’ve had released for months now?

JC 🇱🇻🤝🇺🇦 e/acc 的头像
JC 🇱🇻🤝🇺🇦 e/acc2 天前

Don't run ai models inside the editor window. If the editor crashes or gets closed - your session will end.

ByteCrafter 的头像
ByteCrafter3 天前

mcp itself is just json-rpc plus tool schemas, so the useful part is which editor operations get exposed. undo is the hard bit. only edits wrapped in the editor's transaction system can be rolled back, and plenty of asset operations aren't

Chyld 的头像
Chyld3 天前

this is epic.

ben 的头像
ben3 天前

@bot remind me to set this up on my pc when i get home

Binoui 的头像
Binoui3 天前

MCP is so last year, make a cli like unity it's much better

ExpLiciTSainT 的头像
ExpLiciTSainT1 天前

Can someone please explain to me why anyone would run a CLI inside unreal engine instead of running a CLI outside of it? Why wouldn’t you just have Claude write c++?

Beep 的头像
Beep3 天前

Step 1 finished. Time for step 2

Stanlin Dias 的头像
Stanlin Dias2 天前

Two design choices in the docs stand out 🎮 Tool search is on by default, so agents see a few meta-tools instead of every schema, and tool calls run serially on the game thread. The first saves context, the second means one agent at a time in the editor.

Lucas 的头像
Lucas3 天前

Move to cli, this is better and more powerful.

Stephen Jayakar 的头像
Stephen Jayakar2 天前

melee in UE5 here we come

runking 的头像
runking2 天前

Most of the UE tooling I built was bridge code, so official support is welcome.

ℙ𝕙𝕚 ℂ𝕒𝕡𝕚𝕥𝕒𝕝 𝕀𝕟𝕧𝕖𝕤𝕥𝕞𝕖𝕟𝕥𝕤 的头像
ℙ𝕙𝕚 ℂ𝕒𝕡𝕚𝕥𝕒𝕝 𝕀𝕟𝕧𝕖𝕤𝕥𝕞𝕖𝕟𝕥𝕤2 天前

😎

Danny Chavez 的头像
Danny Chavez3 天前

super interesting, I am going to give a try and will see the result

Flora 🌼 的头像
Flora 🌼3 天前

❤️😍

相关视频

New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic Anthropic and taught by Elie Schoppik Elie Schoppik, its Head of Technical Education. Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on. MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely. In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers. Here’s what you’ll do: - Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources - Learn the core components of the client-server architecture of MCP and the underlying communication mechanism - Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application - Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector - Create an MCP client inside your chatbot to dynamically connect to your server - Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown - Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients - Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications - Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work. Please sign up here!

Andrew Ng

142,256 次观看 • 1 年前