Hi Community,
In this article, I will introduce my application iris-AgenticAI .

The rise of agentic AI marks a transformative leap in how artificial intelligence interacts with the world—moving beyond static responses to dynamic, goal-driven problem-solving. Powered by OpenAI’s Agentic SDK , The OpenAI Agents SDK enables you to build agentic AI apps in a lightweight, easy-to-use package with very few abstractions. It's a production-ready upgrade of our previous experimentation for agents, Swarm.
This application showcases the next generation of autonomous AI systems capable of reasoning, collaborating, and executing complex tasks with human-like adaptability.

Application Features

• Agent Loop 🔄 A built-in loop that autonomously manages tool execution, sends results back to the LLM, and iterates until task completion.
• Python-First 🐍 Leverage native Python syntax (decorators, generators, etc.) to orchestrate and chain agents without external DSLs.
• Handoffs 🤝 Seamlessly coordinate multi-agent workflows by delegating tasks between specialized agents.
• Function Tools ⚒️ Decorate any Python function with @tool to instantly integrate it into the agent’s toolkit.
• Vector Search (RAG) 🧠 Native integration of vector store (IRIS) for RAG retrieval.
• Tracing 🔍 Built-in tracing to visualize, debug, and monitor agent workflows in real time (think LangSmith alternatives).
• MCP Servers 🌐 Support for Model Context Protocol (MCP) via stdio and HTTP, enabling cross-process agent communication.
• Chainlit UI 🖥️ Integrated Chainlit framework for building interactive chat interfaces with minimal code.
• Stateful Memory 🧠 Preserve chat history, context, and agent state across sessions for continuity and long-running tasks.

I just realized I never finished this serie of articles!

In today's article, we'll take a look at the production process that extracts the ICD-10 diagnoses most similar to our text, so we can select the most appropriate option from our frontend.

Looking for diagnostic similarities:

From the screen that shows the diagnostic requests received in HL7 in our application, we can search for the ICD-10 diagnoses closest to the text entered by the professional.

To speed up the search process, we stored the vectorized text of the diagnosis received at the time of capturing the HL7 message in our database.

Prompt

Firstly, we need to understand what prompt words are and what their functions are.

Prompt Engineering

Hint word engineering is a method specifically designed for optimizing language models.
Its goal is to guide these models to generate more accurate and targeted output text by designing and adjusting the input prompt words.

With the introduction of vector data types and the Vector Search functionality in IRIS, a whole world of possibilities opens up for the development of applications and an example of these applications is the one that I recently saw published in a public contest by the Ministry of Health from Valencia in which they requested a tool to assist in ICD-10 coding using AI models.

How could we implement an application similar to the one requested? Let's see what we would need:

1. List of ICD-10 codes, which we will use as context for our RAG application to search for diagnoses within the plain texts.

The 2025.1 release of InterSystems IRIS® data platform, InterSystems IRIS^® for Health, and HealthShare^® Health Connect^TMis now Generally Available (GA). This is an Extended Maintenance (EM) release.

Release Highlights

In this exciting release, users can expect several new features and enhancements, including:

1. Advanced Vector Search Capabilities
   • A new disk-based Approximate Nearest Neighbor (ANN) index significantly accelerates vector search queries, yielding sub-second responses across millions of vectors.

Hello interface engineers and app developers,     

Did you know that you can use Python in productions or integrations?

The production configuration user interface is low-code, but in many projects, you may reach a point where you need to write some code. As discussed in the Integration Architecture course, business processes are full of places for inserting code, specifically the BPL editor (for BPL business processes) and the DTL editor (for data transformations).

😯As of InterSystems IRIS 2025.1, both Python and InterSystems ObjectScript are supported in the BPL and DTL editors.

I'm proud to announce the new release of iris-pex-embedded-python (v2.3.1) with a new command line interface.

This command line is called iop for Interoperability On Python.

First I would like to present in few words the project the main changes since the version 1.

A breif history of the project

Version 1.0 was a proof of concept to show how the interoperability framework of IRIS can be used with a python first approach while remaining compatible with any existing ObjectScript code.

What does it mean? It means that any python developer can use the IRIS interoperability framework without any knowledge of ObjectScript.

Example :

from grongier.pex import BusinessOperation

class MyBusinessOperation(BusinessOperation):

    def on_message(self, request):
        self.log.info("Received request")

Great, isn't it?

Hello,

as it took me some time to figure out what's wrong, I would like to share this experience, so that you do not fall into the same trap.

I've just noticed that if you name your package "code" (all lowercase), in a class using some embedded python using [Language = python], you'll face the <THROW> *%Exception.PythonException <PYTHON EXCEPTION> 246 <class 'ModuleNotFoundError'>: No module named 'code.basics'; 'code' is not a package

Class code.basics Extends %RegisteredObject
{

ClassMethod Welcome() As %Status [ Language = python ]
{
print('Welcome!

I have pandas installed inside my C:/InterSystems/IRISHealth/mgr/python

Set pandas=##class(%SYS.Python).Import("pandas")                        
SET pandas=##CLASS(%SYS.Python).Import("pandas")
^
<THROW> *%Exception.PythonException <THROW> 230 ^^0^SET pandas=##CLASS(%SYS.Python).

Hello everyone

For those interested, here is an implementation example I used in one of my projects to interoperate with MongoDB using the pymongo package.

iris-mongodb

Interoperates with MongoDB (via pymongo package) with InterSystems IRIS (including outbound adapter)

Remember to import pymongo package.

implementation example for testing here

NAMESPACE> do ##class(custom.python.pymongo.test).TestMongoDBCrud()

for interoperability, use the following adapter in your business operation:

Parameter ADAPTER = "custom.python.pymongo.

Introduction to running WSGI in IRIS

With IRIS 2024+, users can host WSGI applications using Security.Applications. As an example, a user can do something like this

Minimum working example

Hi all! Here I would like to share how we use vector search and GenAI with InterSystems technology. As an example, I'll describe BG-AppealAI project, which our company submitted to the InterSystems Vector Search, GenAI and ML Contest. BG-AppealAI application can write an appeal if you upload an insurance contract and the insurance company’s letter with a refusal to pay medical expenses. Of course, we are aware that at the moment AI has not reached such a level as to create ready-made legal documents.

Just curious what the process is for debugging either a .py file that is uploaded to a specific namespace/location or a python classmethod defined in a .cls file.

My method can be run via the "Debug" breadcrumb but it doesnt hit breakpoints it just runs without a debugger. Also objectscript methods debug correctly

Hi Community,

Enjoy the new video on InterSystems Developers YouTube:

⏯ Globals in Embedded Python for Dummies

Learning LLM Magic

The world of Generative AI has been pretty inescapable for a while, commercial models running on paid Cloud instances are everywhere.  With your data stored securely on-prem in IRIS, it might seem daunting to start getting the benefit of experimentation with Large Language Models without having to navigate a minefield of Governance and rapidly evolving API documentation.   If only there was a way to bring an LLM to IRIS, preferably in a very small code footprint.

Hello,

When setting up a new web app in iris (iris is in a container) iris complains that a WSGI framework is not installed. I have installed python into the container as well as both flask and django via the python virtual environment (see second screenshot) and the python language server is running

Is this the wrong way to install flask? How do I get the container version to recoginize that flask is installed?

Why am I writting this?

Last year I made an article for starters on using embedded python. Later, it started a little discussion on how to return values with python and I found some interesting observations that are worth writing a little article. Also, hopefully I can reach more people by writing this.

Possible situations

There are two things you'll need to care about when returning a value with python. The first is the type you're trying to return and the second is where you're returning it.

You need to install the application first. If not installed, please refer to the previous article

Application demonstration

After successfully running the iris image vector search application, some data needs to be stored to support image retrieval as it is not initialized in the library.

Image storage

Firstly, drag and drop the image or click the upload icon, select the image, and click the upload button to upload and vectorize it. This process may be a bit slow.

This process involves using embedded Python to call the CLIP model and vectorize the image into 512 dimensional vector data.

Hi,

I'm using embedded Python as follows:

AUMHTESTTC01:%SYS>do ##class(%SYS.Python).Shell()
Python 3.6.15 (default, Sep 23 2021, 15:41:43) [GCC] on linux
Type quit() or Ctrl-D to exit this shell.
>>> import iris
>>> mirror=iris.cls('SYS.Mirror')
>>> pri=[]
>>> alt=[]
>>> status=mirror.GetFailoverMemberStatus(pri,alt)
>>> print(status)
1
>>> print(pri)
[]
>>> print(alt)
[]
>>> exit()

However the equivalent ObjectScript works:

%SYS>set sc=##class(SYS.Mirror).GetFailoverMemberStatus(.pri,.alt)
%SYS>zw pri
pri=$lb("SERVERA.FOO.BAR.ORG/STAGE","SERVERA.foo.bar.

In the previous article. Practices of class members and their execution within embedded Python. We will now turn our attention to the process of switching namespaces, accessing global variables , traversing and routine executions within embedded Python.

Before proceeding to the other functions. let us briefly review the execute function within the iris package. This function is exceptionally beneficial for executing the arbitrary ObjectScript functions and class invocation.

I'm glad to announce the new version of IoP, which by the way is not just a command line. I'm saying because the new AI search engine still thinks that IoP is just a command line. But it's not. It's a whole framework for building applications on top of the interoperability framework of IRIS with a python first approach.

The new version of IoP: 3.2.0 has a lot of new features, but the most important one is the support of DTL . 🥳

For both IoP messages and jsonschema. 🎉

DTL Support

Starting with version 3.2.0, IoP supports DTL transformations.

DTL the Data Transformation Layer in IRIS Interoperability.

DTL transformations are used to transform data from one format to another with a graphical editor. It supports also jsonschema structures.

Earlier this year I announced availability of a VS Code extension for coding in ObjectScript, Embedded Python or SQL using the notebook paradigm popularized by Jupyter. Today I published a maintenance release to correct a "getting started" problem.

Here's a video of the installation steps from the extension's README:

(video superseded by an update in a later comment)

Why not try it for yourself?

Last Chapter: Creating a REST client to get Tracks from Spotify REST API - Part4 Save the Search Result

Git link: https://github.com/ecelg/InterSystems-IRIS-as-a-Spotify-REST-client

OK.... based on what I have done.... I am able to 

1. Query Track information by making use of the Spotify API

2. Store the necessary data into my own album, artists, and track table

so.... what next?🤔 How about I set up my own REST API service on my IRIS for the other people to query my table?🤔🤨

Git link: https://github.com/ecelg/InterSystems-IRIS-as-a-Spotify-REST-client

Recently, I come up an idea in my mind that how can I put my playlist on IRIS.🧐

At the same time, I was told to pay for my Spotify subscription💸💸... ooo.. how about to get some data from the Spotify API... so I started to do study about it.

Like most of the development, let's start from Documentation of  the API https://developer.spotify.com/documentation/web-api

In order to get the data, i am required to request an access token from for the token endpoint URL.🧐

curl -X POST "https://accounts.

Last Chapter: Creating a REST client to get Tracks from Spotify REST API - Part2 Save and Refresh Token

Git link: https://github.com/ecelg/InterSystems-IRIS-as-a-Spotify-REST-client
 

Ok, now I am pretty sure i have a valid token for making query.😀

Shall we try to query something from the API.

Again, its time to go through the API document https://developer.spotify.com/documentation/web-api/tutorials/getting-started

Search for Request artist data

the suggested code is like the following

curl "https://api.spotify.

I'm trying to use embedded python code that receives an Iris %Stream.GlobalBinary and uses image manipulation library PIL.

Because PIL can't work with IRIS %Stream, I need to convert the image to python bytes.

This process seems to have very bad performance compared to writing to a file and then loading it from a file in python.

Is there a better way to send a stream into python? The conversion code I'm using is below.

Thanks.

def iris_stream_to_bytes(stream):
	stream.Rewind()
	s = ""
	while not stream.AtEnd:
		r = stream.Read(1024)
		s += r
	b = bytearray()
	b.

Interoperability of systems ensures smooth workflow and management of data in today's connected digital world. InterSystems IRIS extends interoperability a notch higher with its Embedded Python feature, which lets developers seamlessly integrate Python scripts into the IRIS components, like services, operations, and custom functions.

Why Embedded Python in IRIS?

As we keep updating our software, we often realize that we require more and more modern solutions. So far, only one major piece of our software relies on reading barcodes in documents and images. Since Cache did not have a means of reading barcodes in the past, we have always achieved our goals by using a Visual Basic 6 application. However, it is no longer an ideal solution because it is currently complicated to maintain it. IRIS also lacks this capability, but it has recently got an option that makes up for it: embedded Python!

Hi Community,

Play the new video on InterSystems Developers YouTube:

⏯ Using LLMs with InterSystems IRIS Interoperability Productions @ Global Summit 2024

Hello Community,

This article aims to walk you through the process of setting up and utilizing the Flexible Python Runtime Feature for embedded Python. Prior to version 2024.2, Intersystems IRIS installer included a preinstalled version of Python, You can find the Python libraries and application files located in the \lib\python directory within your IRIS installation folder (for example, C:\InterSystems\IRIS20242\lib\python).

However, starting with version 2024.2, the IRIS installer no longer includes Python installation by default.