>>> [!note] Migrated issue <!-- Drupal.org comment --> <!-- Migrated from issue #3581443. --> Reported by: [scott falconer](https://www.drupal.org/user/52557) >>> <p>This issue was spun off from <span class="drupalorg-gitlab-issue-link project-issue-status-info project-issue-status-1"><a href="https://www.drupal.org/project/drupal/issues/3574093" title="Status: Active">#3574093: Ban slop issue summaries and comments</a></span> to focus only on responsible LLM use training and guidance for Drupal contributors.</p> <p>Please keep discussion here scoped to training, enablement, and tools. This issue should build on policy decisions as documented in <a href="https://www.drupal.org/docs/develop/issues/issue-procedures-and-etiquette/policy-on-the-use-of-ai-when-contributing-to-drupal" title="Policy on the use of AI when contributing to Drupal">Policy on the use of AI when contributing to Drupal</a>.</p> <p><strong>Problem/Motivation<br> </strong>As Drupal allows some forms of AI-assisted contribution as defined in <a href="https://www.drupal.org/docs/develop/issues/issue-procedures-and-etiquette/policy-on-the-use-of-ai-when-contributing-to-drupal" title="Policy on the use of AI when contributing to Drupal">Policy on the use of AI when contributing to Drupal</a>, contributors need practical training and tools to use these capabilities responsibly, and without shifting the review burden onto other contributors, or negatively impacting those who choose not the use LLMs.</p> <p>Existing contributor and review-guide work is helpful, but there is still room for a more explicit training path focused on responsible LLM use in Drupal contribution workflows.</p> <p><strong>Proposed resolution<br> </strong>Create training/guidance materials and tools for responsible LLM use in Drupal contribution workflows.</p> <p><strong>Possible topics:<br> </strong></p> <ul> <li>Understanding the issue and recent discussion before prompting</li> <li>Disclosure expectations</li> <li>Local testing and verification</li> <li>Explaining generated output during review</li> <li>Revising code without re-dumping AI output</li> <li>Respecting maintainer and reviewer time</li> <li>When not to use AI</li> <li>Mentorship / onboarding implications</li> </ul> <p><strong>Possible outputs:<br> </strong></p> <ul> <li>contributor documentation</li> <li>AI generation best practices</li> <li>sample AGENTS.md files, Agent skills, and tooling</li> <li>workshop/session materials</li> <li>checklist or playbook</li> <li>Dependencies</li> </ul> <p>AI disclosure, the above was researched and discussed with AI prior to posting, it was then reviewed and drafted by humans. </p> > Related issue: [Issue #3574093](https://www.drupal.org/node/3574093) > Related issue: [Issue #3565917](https://www.drupal.org/node/3565917) > Related issue: [Issue #3580299](https://www.drupal.org/node/3580299) > Related issue: [Issue #3557005](https://www.drupal.org/node/3557005) > Related issue: [Issue #3569240](https://www.drupal.org/node/3569240) > Related issue: [Issue #3580761](https://www.drupal.org/node/3580761)