From b42aada719dcec7c398953ebe82f10549c397444 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Juan=20Antonio=20Bre=C3=B1a=20Moral?= Date: Mon, 9 Mar 2026 08:27:53 +0100 Subject: [PATCH] Improving the article --- docs/blog/2026/03/release-0.12.0.html | 4 ++-- docs/feed.xml | 2 +- site-generator/content/blog/2026/03/release-0.12.0.md | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/docs/blog/2026/03/release-0.12.0.html b/docs/blog/2026/03/release-0.12.0.html index f55f2cb7..c70b0d88 100644 --- a/docs/blog/2026/03/release-0.12.0.html +++ b/docs/blog/2026/03/release-0.12.0.html @@ -545,7 +545,7 @@

Improvements in System prompts

  • Added support to generate AGENTS.md with @173-java-agents: Added this specific missing feature.

    • Conclusions

    -

    If you followed the article, the project is evolving from System prompts to Skills, but during the time between v0.11.0 to v0.12.0, new elements have appeared in the market, such as: Subagents, Commands, Hooks, Plugins, Spec-driven. The project will continue evolving its current scope plus Subagents.

    +

    If you followed the article, the project is evolving from System prompts to Skills—and continues to advance practices such as curated shared instructions for software teams and AGENTS.md, both recognized in the Thoughtworks Technology Radar. During the time between v0.11.0 to v0.12.0, new elements have appeared in the market, such as: Subagents, Commands, Hooks, Plugins, Spec-driven. In the next release, the project will review how to use Skills with Subagents.

    Source: https://code.claude.com/docs/en/agent-teams

    When you talk with people using these technologies, you notice that software delivery has increased—but design and requirements gathering must keep pace on quality too. If you accelerate delivery without investing in requirements, you end up with fast products that conceal potential design flaws.

    @@ -555,7 +555,7 @@

    Conclusions

    EA, PO, BA, SA, TL | TL, SWE | TL, SWE, SE -

    Reviewing in perspective, this kind of products has improved so much the performance, now you can pick a complex feature and if the requirements are complete, the execution using techniques like Plan mode will help so much in the execution without the bad experience from the past (Madrid Jug - May 2025).

    +

    Putting this in perspective, this kind of product has improved performance so much that you can now pick a complex feature and, if the requirements are complete, execution using techniques like Plan mode will help greatly—without the poor experience from the past (Madrid Jug - May 2025). Using Plan Mode doesn't require you to change how you organize your requirements, unlike new SDD tools such as Spec-kit.

    Do you still have questions about the project?

    If you feel stuck using this project or have questions, you can attend the following Workshop at Codemotion Madrid 2026:

    diff --git a/docs/feed.xml b/docs/feed.xml index 87c7047d..314f3aff 100644 --- a/docs/feed.xml +++ b/docs/feed.xml @@ -2,7 +2,7 @@ https://jabrena.github.io/cursor-rules-java/ Cursor rules and Skills for Java - 2026-03-08T18:06:37+0100 + 2026-03-09T08:27:24+0100 diff --git a/site-generator/content/blog/2026/03/release-0.12.0.md b/site-generator/content/blog/2026/03/release-0.12.0.md index 9df2ef8b..3756edba 100644 --- a/site-generator/content/blog/2026/03/release-0.12.0.md +++ b/site-generator/content/blog/2026/03/release-0.12.0.md @@ -576,7 +576,7 @@ Since the previous version, the project has invested time improving the current ## Conclusions -If you followed the article, the project is evolving from `System prompts` to `Skills`, but during the time between [`v0.11.0`](https://github.com/jabrena/cursor-rules-java/releases/tag/0.11.0) to [`v0.12.0`](https://github.com/jabrena/cursor-rules-java/releases/tag/0.12.0), new elements have appeared in the market, such as: `Subagents`, `Commands`, `Hooks`, `Plugins`, `Spec-driven`. The project will continue evolving its current scope plus **Subagents**. +If you followed the article, the project is evolving from `System prompts` to `Skills`—and continues to advance practices such as curated shared instructions for software teams and `AGENTS.md`, both recognized in the [Thoughtworks Technology Radar](https://www.thoughtworks.com/radar/techniques). During the time between [`v0.11.0`](https://github.com/jabrena/cursor-rules-java/releases/tag/0.11.0) to [`v0.12.0`](https://github.com/jabrena/cursor-rules-java/releases/tag/0.12.0), new elements have appeared in the market, such as: `Subagents`, `Commands`, `Hooks`, `Plugins`, `Spec-driven`. In the next release, the project will review how to use `Skills` with **Subagents**. ![](/cursor-rules-java/images/2026/3/subagents.png) @@ -592,7 +592,7 @@ Jira / Github / Azure Devops / RedMine > Plan Mode > Agent Mode to deliver EA, PO, BA, SA, TL | TL, SWE | TL, SWE, SE ``` -Reviewing in perspective, this kind of products has improved so much the performance, now you can pick a complex feature and if the requirements are complete, the execution using techniques like `Plan mode` will help so much in the execution without the bad experience from the [past (Madrid Jug - May 2025)](https://jabrena.github.io/101-cursor/v010/index.html#/10/15). +Putting this in perspective, this kind of product has improved performance so much that you can now pick a complex feature and, if the requirements are complete, execution using techniques like `Plan mode` will help greatly—without the poor experience from the [past (Madrid Jug - May 2025)](https://jabrena.github.io/101-cursor/v010/index.html#/10/15). Using `Plan Mode` doesn't require you to change how you organize your requirements, unlike new SDD tools such as Spec-kit. ## Do you still have questions about the project?