DEV Community: Derek

Tutorial Codex/Claude + Craftsman Agent Skills Becomes the AI 3D Toy IP Generator Workstation Turn Your IPs to Digital Assets and Finished Products

Derek — Fri, 28 Aug 2026 03:22:34 +0000

Hi Dev community, I this blog I will introduce how to use your local Desktop Agent and Craftsman Agent API/Skills as a unified AI 3D Toy IP Generator Workstation. You can turn your creative ideas or IP into ready to produce Toys (AI Figurine,Plush Stuffed Toys), multiview sheets, downloadable 3D models and more.

Requirements:

Your Local Agents Codex/Claude/Gemini CLIs/Workbuddy
Craftsman Agent Skills, Especially the toy_generator_design_draft,toy_generator_task_create,toy_generator_task_poll
OneKey Agent Gateway Access Key: DEEPNLP_ONEKEY_ROUTER_ACCESS Required to use the SOTA Image Generation (Nano Banana/Image Gen ) and 3D models APIs (Tripo/Meshy) in the toy design agentic workflow which can be access using OneKey MCP CLI or APIs.

And you can generate:

Multiview sheet of your IPs.
3D models(.obj,.glb) ready for 3D printing and Plush Toy Manufacturing on Craftsman Agent Store, which is the manufacturing on Demand platform to connect to 1000+ craftsman workshop and turn your design to finished products and delivery.

Online Web:
AI Toy Generator Craftsman Agent Web
Craftsman Agent Skills GitHub

Here is a gallery of demo generated Plush Toy Steve in Minecraft, PVC Materials AI Figurine Steve/Alex/Kulipa, etc.

Example 1, Minecraft IPs

Example 2, Swimming Boys and Girls

Example 3, Cult Movie Niulai

This is the share workstation online view of Niulai and 3D models:
https://craftsman-agent.aiagenta2z.com/app/sessions/share/296aaac1-14f5-4c85-989e-cd69e917a5eb

Tutorial

The typical workflow of IP design and 3D/Toy Generation is:

[Text/Image Prompt]
        ->
[Toy Design Draft / Multi-View Sheet]
        ->
[Create 3D Generation Task: Task ID]
        ->
[Poll Task Results Progress using Task ID]
        ->
[3D Toy Model + Preview]

Setup Agent Skills Craftsman Agent

In your local project folder, setup the skills markdown file by running below commands to download skills from GitHub or Clawhub

npx skills add ai-hub-admin/craftsman-agent

openclaw skills install @ai-hub-admin/toy-generator

You can download the toy-generator and more specific usage from the hub. The raw skills data can be found in GitHub Repo

https://github.com/AI-Hub-Admin/Craftsman-Agent

2. Running Skills

2.1 Generate Toy 3D Model MultiView Sheets Images

Example: I am using the Niulai (A Cow) Cult Movie for low resolution animation screenshots of videos images are reference.
Niulai Movie Cow Picture : https://craftsman-agent.aiagenta2z.com/static/ogino-chihiro/50923d3d-8222-45ce-989e-4c41ec15e3ab/3fd95298d85b42389415cd8d5f8227cd.jpg

You can be specific about which kind of toys you would like, such as the Plush Stuffed Animal Toys or AI Figurine (PVC materials), 3D Printing Materials, etc.
You can also define the texture of the toys to make it better

Setup the Env Variables before Running the Skills APIs, Otherwise you will an Key not found error! You can get the onekey access variables and demo credits at https://deepnlp.org/workspace/keys.

export DEEPNLP_ONEKEY_ROUTER_ACCESS=your_access_key

This is the final Prompt

Please follow the reference pictures of the Movie Niulai of a standing Cow. Please Output a multiview sheets images of stuffed toy. 
Background should be white. Facial Expression should resemble the main face. Texture of the cow should be exact like the main picture, coarse grain texture.
Use the raw character images as references: https://craftsman-agent.aiagenta2z.com/static/ogino-chihiro/50923d3d-8222-45ce-989e-4c41ec15e3ab/3fd95298d85b42389415cd8d5f8227cd.jpg

The Output of the MultiView Sheets will give the Front View, Side View and Back View of the toys when the skills runs

Multi View Sheets of Craftsman Agent Niulai

This file will be ready to send to Manufacturing Supplier to make it out and the factory will print out the design, get modules of plush materials and make Demo Toys!
Just upload the generated images to marketplace, and make temperary order to find Craftsman and your order will be matched, https://craftsman-agent.aiagenta2z.com/marketplace

2.2 Generate 3D Models for 3D Printing or AI Figurine

If you want to use 3D Printing to print out the model and paint. You can try below prompts and output a 3D models calls
the default SOTA 3D generation APIs (Tripo/Meshy/etc). Note that the 3D generate have 3 models (image-to-model, text-to-model, multi-view sheets to model)

In this blog we use the outputs of previous step 2.1 MultiView Sheets and produce a more accurate models. Alternatively, you can also
use just single images to generate (But the back and side e.g. the tail of the cow might be different than your expectations.)

Use Craftsman Agent Toy Generator Skills to Generate 3D Models of Niulai. Use the previously generated MultiView Sheets Front, Side and Back Images to generate 3D Models ready to ship.

Note that the 3D Generate Takes Minutes the process work like this:

Toy 3D Generation Start -> Generate Task ID task_id -> Polling on Task ID task_id progress to get final results

Get the Task View From the First API Calls

From the First API calls Get the 3D ShareURL Views

Field	Description
session_id	The generation unique session id
share_url	To View the Generated Images or Slides in Workspace

The is a final generated gallery of 3D models Polling:
https://craftsman-agent.aiagenta2z.com/app/sessions/share/296aaac1-14f5-4c85-989e-cd69e917a5eb

A Demo URL Path of Generated 3D models Ready to Download: (Path are marked in Tutorial and live for a lifecycle)

https://us-static.aiagenta2z.com/xxxxxxxx/user_name/session_id/tripo_pbr_model_990f448b-d8b8-491e-8d9f-284802aa985c.glb

And This is the final Toy generated 3d files ready to print/manufacture!

https://craftsman-agent.aiagenta2z.com/app/sessions/share/296aaac1-14f5-4c85-989e-cd69e917a5eb

LangChain x AI Agent A2Z Agent Deployment Tutorial on How to Bring Agent Live

Derek — Tue, 17 Feb 2026 03:23:31 +0000

Abstract

Deploy the framework based AI Agent online is always difficult and in this blog, we will introduce
how to deploy a langchain framework based AI agent and bring it from local to online service with an /chat endpoint. Tutorials covers two examples content-builder-agent and deep_research in LangChain DeepAgents Repo(github: langchain-ai/deepagents) and the Agent Deployment templates GitHub can be found and easily deployed on A2Z Deployment A2Z Deployment Doc and A2Z Deployment Platform. After deployment, we can bring the agent live and get "/chat" endpoints.

Tutorial

Step 1. Convert the LangChain DeepAgents to LiveRunTime

The agent class in their original implementation has two skills blog-post and social-media
and is created using the create_deep_agent base function.

def create_content_writer():
    """Create a content writer agent configured by filesystem files."""
    return create_deep_agent(
        memory=["./AGENTS.md"],           # Loaded by MemoryMiddleware
        skills=["./skills/"],             # Loaded by SkillsMiddleware
        tools=[generate_cover, generate_social_image],  # Image generation
        subagents=load_subagents(EXAMPLE_DIR / "subagents.yaml"),  # Custom helper
        backend=FilesystemBackend(root_dir=EXAMPLE_DIR),
    )

Step 2. Create a BaseLiveRuntime and implement an Async Generator

The BaseLiveRuntime object produces a FastAPI app which expose and /chat endpoint that takes messages format input.
To make the agent online, the BaseLiveRuntime takes two variables, the first one is the agent object defined by various
framework, such as LangChain/CrewAI/OpenAI Agent SDK/etc, the second one is an async generator which defined how the
agent will run the input, such as agent.run, agent.invoke and customized function.

Define Runtime

from ai_agent_marketplace.runtime.base import *

async def content_builder_stream_generator(
    agent: Any,
    user_query: str,
    **kwargs
) -> AsyncGenerator[str, None]:
    """
    """
    ## more

runtime = BaseLiveRuntime(
    agent=agent,
    stream_handler=content_builder_stream_generator
)

## Returned a FastAPI based app with /chat endpoint
app = runtime.app

Create a Streaming Adapter

Define an async generator that adapts your LangChain agent output into streaming chunks.

The async generator takes in two parameters: agent an customized agent object, user_query
that are parsed from the messages object from the "\chat" endpoints.
In the async generator, the agent calls agent.invoke({"messages": messages}) methods.

from ai_agent_marketplace.runtime.base import *
from typing import Any, AsyncGenerator
import json
import uuid
import asyncio

async def deepagents_stream_generator(
    agent: Any,
    user_query: str,
    **kwargs
) -> AsyncGenerator[str, None]:
    """
    Universal async adapter for LangChain agent
    """

    # Send initial streaming message
    initial_content = "Task Started and Research Take a Few Minutes"
    initial_chunk = json.dumps(
        assembly_message(
            type=MESSAGE_TYPE_ASSISTANT,
            format=OUTPUT_FORMAT_TEXT,
            content=initial_content,
            content_type=CONTENT_TYPE_MARKDOWN,
            section=SECTION_ANSWER,
            message_id=str(uuid.uuid4()),
            template=TEMPLATE_STREAMING_CONTENT_TYPE,
        )
    )

    yield initial_chunk + STREAMING_SEPARATOR_DEFAULT
    await asyncio.sleep(0)

    try:
        # Call LangChain agent
        messages = [{"role": "user", "content": user_query}]
        result = agent.invoke({"messages": messages})

        output_messages = result["messages"] if "messages" in result else []

        for message in output_messages:
            message_id, content, role = extract_message_content_langchain(message)

            output_chunk = json.dumps(
                assembly_message(
                    type=MESSAGE_TYPE_ASSISTANT,
                    format=OUTPUT_FORMAT_TEXT,
                    content=content,
                    content_type=CONTENT_TYPE_MARKDOWN,
                    section=SECTION_ANSWER,
                    message_id=message_id,
                    template=TEMPLATE_STREAMING_CONTENT_TYPE,
                )
            )

            yield output_chunk + STREAMING_SEPARATOR_DEFAULT

    except Exception:
        yield json.dumps({}) + STREAMING_SEPARATOR_DEFAULT

Step 3. Deploy the Agent Live

Go to the deployment workspace (DeepNLP AI Agent A2Z Deployment)

Choose Github Tab
Public url: https://github.com/aiagenta2z/agent-mcp-deployment-templates
Entry Point Command shell

uvicorn langchain_deepagents.deep_research.research_agent_server:app

Set the Environment Variables

# Set API keys
export GOOGLE_API_KEY="..."      # For image generation
export TAVILY_API_KEY="..."      # For web search (optional)

Step 5. Click Deploy and You will get the URL

Get the Product /chat POST URL :

https://langchain-ai.aiagenta2z.com/content-builder-agent/chat

Architecture Summary

LangChain Agent
        ↓
Streaming Adapter (Async Generator)
        ↓
BaseLiveRuntime
        ↓
FastAPI App (/chat)
        ↓
Streaming JSON Response

Step 4. Test Deployed Agent with curl

Case 1: Simple Math

curl -X POST "http://localhost:8000/chat" \
  -H "Content-Type: application/json" \
  -d '{"messages":[{"role":"user","content":"Calculate 1+1 result"}]}'

Sample Streaming Output

{"type":"assistant","format":"text","content":"Task Started...","section":"answer","message_id":"670d3458-a539-406f-a786-1afc0f0fc201","content_type":"text/markdown","template":"streaming_content_type"}
{"type":"assistant","format":"text","content":"Calculate 1+1 result","section":"answer","message_id":"701be311-37e3-4ee1-9519-6d8e65b47f59","content_type":"text/markdown","template":"streaming_content_type"}
{"type":"assistant","format":"text","content":"1 + 1 = 2","section":"answer","message_id":"lc_run--019c55fe-4ed2-7da3-9e05-0a8758aa10cc-0","content_type":"text/markdown","template":"streaming_content_type"}

Case 2: Research Task

curl -X POST "http://localhost:8000/chat" \
  -H "Content-Type: application/json" \
  -d '{"messages":[{"role":"user","content":"research context engineering approaches used to build AI agents"}]}'

Sample Streaming Output (Truncated)

{"type":"assistant","content":"Task Started..."}
{"type":"assistant","content":"Updated todo list ..."}
{"type":"assistant","content":"Updated file /research_request.md"}
{"type":"assistant","content":"Here is a comprehensive report on context engineering approaches..."}

The response is streamed incrementally as the agent reasons, calls tools, and produces final output.

Deploy And Test Examples

You can also deploy publicly:

curl -X POST "https://deepagents.aiagenta2z.com/deep_research/chat" \
  -H "Content-Type: application/json" \
  -d '{"messages":[{"role":"user","content":"Calculate 1+1 result"}]}'

{"type":"assistant","content":"Task Started..."}
{"type":"assistant","content":"Updated todo list ..."}
{"type":"assistant","content":"Updated file /research_request.md"}
{"type":"assistant","content":"Here is a comprehensive report on context engineering approaches..."}

DEV Community: Derek

Tutorial Codex/Claude + Craftsman Agent Skills Becomes the AI 3D Toy IP Generator Workstation Turn Your IPs to Digital Assets and Finished Products

Requirements:

And you can generate:

Example 1, Minecraft IPs

Example 2, Swimming Boys and Girls

Example 3, Cult Movie Niulai

Tutorial

2. Running Skills

2.1 Generate Toy 3D Model MultiView Sheets Images

Multi View Sheets of Craftsman Agent Niulai

2.2 Generate 3D Models for 3D Printing or AI Figurine

Get the Task View From the First API Calls

Related Reading and Usage

LangChain x AI Agent A2Z Agent Deployment Tutorial on How to Bring Agent Live

Tutorial

Step 1. Convert the LangChain DeepAgents to LiveRunTime

Step 2. Create a BaseLiveRuntime and implement an Async Generator

Define Runtime

Create a Streaming Adapter

Step 3. Deploy the Agent Live

Architecture Summary

Step 4. Test Deployed Agent with curl

Case 1: Simple Math

Sample Streaming Output

Case 2: Research Task

Sample Streaming Output (Truncated)

Deploy And Test Examples