#Generative AI (GenAI)

In last post I talked about iris-copilot, an apparent vision that in near future any human language is a programming language for any machines, systems or products. Its agent runners were actually using such so-called 3rd generation of agents. I want to keep/share a detailed note on what it is, for my own convenience as well. It was mentioned a lot times in recent conversations that I was in, so probably worth a note.

I'm starting to play more with AI enabled coding. 
I've been using Github Copilot inside Visual studio code, which is very good at coming up with autocomplete suggestions that are accurate and useful. (Along with some utter rubbish, naturally).
For web development I'm starting to use Claude Code in VS Code to help create web sites and integrations. I want to see how it can help with IRIS development. 

However I can't get claude to read any iris code directly as I'm connected to my server via isfs server connections.

Over the past year there have been a few DC articles offering MCP servers designed to connect to IRIS and help the AI features of VS Code and its cousins do a better job. For example:

• Hey chat, what's up with my Interoperability by @Dmitry Maslennikov 
• No-Hallucinate MCP: Real-time Routine Context for ObjectScript by @Jinyao 
• iris-mcp-atelier introduced by @Ernest Salas in his comment on the Claude Code vs ISFS. Has anyone cracked how to get one to see the other? post by @MarkCharlton.

Motivation

Why do we need this?

1. Lack of Compiled Context: AI tools only see source code; they don't know what the final compiled routine looks like.

2. Macro Hallucination: Because AI doesn't see our #include files or system macros, it often makes them up, wasting time during debugging.

3. The Documentation Gap: Deep logic optimization often requires understanding internal macros that aren't fully covered in public documentation.

4. Manual Overhead: Currently, the only way to fix this is to manually use the IRIS VS Code extension to find the "truth" in the routine.

v2026.1 was just released as GA, and one of the features I'm looking forward to using is the DTL Explainer feature.

This allows you to take a Data Transformation, and with a click of a button get a human-readable description of the transformation (which you can also use as the basis for the DTL Description).

For complex DTLs, especially ones you didn't write yourself, or you did but a long time ago, this will allow you to get a clear quick understanding of what it's doing.