---
title: "Publication on Habr: history of prompts to AI agents"
description: "In the Habr article I note that the main artifact of development with AI agents becomes the instructions and the history of prompts. In a poll, 50% of readers confirmed that they save such iterations."
date: 2025-07-15
display_date: "July 15, 2025"
image: "https://habrastorage.org/r/w1560/getpro/habr/upload_files/acd/bc7/ec1/acdbc7ec1b608f00e5637520c6301071.png"
image_alt: "Publication on Habr: history of prompts with AI agents"
---

# Publication on Habr: history of prompts to AI agents

<zoom-img
            src="https://habrastorage.org/r/w1560/getpro/habr/upload_files/acd/bc7/ec1/acdbc7ec1b608f00e5637520c6301071.png"
            alt="Publication on Habr"
            width="100px"
            style="float: left; margin: 15px 15px 0 0;"
    ></zoom-img>

My article [“Do you save the history of prompts to AI agents?”](https://habr.com/ru/articles/927966/) has been published on the [Habr](https://habr.com/) platform.

In the article, I share my experience of using OpenAI Codex and conclude that when developing with AI agents, the key artifact is not so much the code or documentation, but **the instructions and the history of prompts**. They reflect the tasks that were set and the results that were obtained.

I added a poll to the publication. Half of the participants (50%) answered that they keep the history of iterations, and another third (33%) found the idea interesting. This shows that keeping a record of interactions with AI is already in demand.