{"description":"Welcome to the best place to find Luanti mods, games, and texture packs","expired":false,"feed_url":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases_feed.json","home_page_url":"https://content.luanti.ru/","icon":"https://content.luanti.ru/favicon-128.png","items":[{"author":{"avatar":"https://secure.gravatar.com/avatar/1ac9d721d5c946f1b26be6dfd99d6a36?s=64&d=retro&r=g","name":"LichenCrown","url":"https://content.luanti.ru/users/roastlichen/"},"content_html":"<p>A command was added that allows receiving responses from ollama via the Windows console. However, this recent implementation is experimental and therefore does not have the same optimizations used in Linux.</p>\n","date_published":"2026-02-25T20:13:32Z","id":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35225/","image":"https://content.luanti.ru/thumbnails/4/beb76ed38c.png","language":"en","summary":"A command was added that allows receiving responses from ollama via the Windows console. However, this recent implementation is experimental and therefore does not have the same optimizations used in Linux.","tags":["release"],"title":"\"ShameGPT\" updated: Windows Implementation (Experimental)","url":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35225/"},{"author":{"avatar":"https://secure.gravatar.com/avatar/1ac9d721d5c946f1b26be6dfd99d6a36?s=64&d=retro&r=g","name":"LichenCrown","url":"https://content.luanti.ru/users/roastlichen/"},"content_html":"<h3 id=\"current-features-and-changes\">Current Features and Changes <a class=\"header-anchor\" href=\"#current-features-and-changes\" rel=\"nofollow\">\ud83d\udd17</a></h3>\n<ul>\n<li>\n<p><strong>Robust Configuration Validation</strong>: The mod checks if <code>model_name</code> is empty or consists only of whitespace, automatically defaulting to <code>gemma3:270m</code> in those cases.</p>\n</li>\n<li>\n<p><strong>Asynchronous Processing</strong>: It uses a <code>globalstep</code> monitor and system-level background execution (<code>&amp;</code>) to prevent the Minetest server from freezing while the AI generates a response.</p>\n</li>\n<li>\n<p><strong>ANSI Escape Sequence Cleaning</strong>: A <code>clean_ai_text</code> function is implemented to strip terminal formatting codes, carriage returns, and leading/trailing whitespace from the AI's output.</p>\n</li>\n<li>\n<p><strong>Security Prompt Filtering</strong>: The <code>is_safe_prompt</code> function uses pattern matching to block dangerous characters (like <code>;</code>, <code>&amp;</code>, <code>|</code>, and <code>$</code>) that could be used for command injection.</p>\n</li>\n<li>\n<p><strong>Binary Presence Check</strong>: The mod utilizes the <code>command_exists</code> function to verify if the <code>ollama</code> executable is available on the system before attempting a request.</p>\n</li>\n<li>\n<p><strong>Request Concurrency Management</strong>: An <code>active_requests</code> table tracks ongoing queries per player to prevent users from spamming the AI service.</p>\n</li>\n<li>\n<p><strong>Automated Cleanup</strong>: Once a response is read, the mod automatically deletes the temporary result and &quot;done&quot; signal files from the system.</p>\n</li>\n</ul>\n<hr>\n<h3 id=\"system-limitations\">System Limitations <a class=\"header-anchor\" href=\"#system-limitations\" rel=\"nofollow\">\ud83d\udd17</a></h3>\n<blockquote>\n<p>[!IMPORTANT]\n<strong>Linux Only</strong>: The current implementation relies on Unix-specific commands and shell operators such as <code>which</code>, <code>touch</code>, <code>2&gt;/dev/null</code>, and the background operator <code>&amp;</code>. Consequently, it <strong>only works on Linux</strong>.\n<strong>Language</strong>: All user-facing messages, command descriptions, and error notifications are currently written <strong>only in English</strong>.</p>\n</blockquote>\n","date_published":"2026-02-21T19:19:33Z","id":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35125/","image":"https://content.luanti.ru/thumbnails/4/beb76ed38c.png","language":"en","summary":"### Current Features and Changes","tags":["release"],"title":"\"ShameGPT\" updated: Error Handling","url":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35125/"},{"author":{"avatar":"https://secure.gravatar.com/avatar/1ac9d721d5c946f1b26be6dfd99d6a36?s=64&d=retro&r=g","name":"LichenCrown","url":"https://content.luanti.ru/users/roastlichen/"},"content_html":"<h3 id=\"-improvements--bug-fixes\">\ud83d\udee0 Improvements &amp; Bug Fixes <a class=\"header-anchor\" href=\"#-improvements--bug-fixes\" rel=\"nofollow\">\ud83d\udd17</a></h3>\n<ul>\n<li>\n<p><strong>Asynchronous AI Processing:</strong> Fixed the critical issue where the game world (chunks) would stop loading and the chat would freeze while waiting for a response from the AI.</p>\n</li>\n<li>\n<p><strong>Non-Blocking Requests:</strong> The mod now utilizes a background process to run Ollama queries, allowing players to continue building and exploring while the chatbot processes their request.</p>\n</li>\n<li>\n<p><strong>Timer-Based Polling:</strong> Integrated a global step timer that checks for AI completion every 0.5 seconds, preventing CPU spikes and ensuring smooth frame rates.</p>\n</li>\n<li>\n<p><strong>Enhanced Security:</strong> Added a prompt validation layer to block forbidden characters (like <code>;</code>, <code>&amp;</code>, or <code>$</code>) to ensure safe command execution.</p>\n</li>\n</ul>\n","date_published":"2026-02-17T23:28:50Z","id":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35070/","image":"https://content.luanti.ru/thumbnails/4/beb76ed38c.png","language":"en","summary":"### \ud83d\udee0 Improvements & Bug Fixes","tags":["release"],"title":"\"ShameGPT\" updated: No More Slow Chunks","url":"https://content.luanti.ru/packages/roastlichen/shamegpt/releases/35070/"}],"title":"ContentDB package updates","version":"https://jsonfeed.org/version/1"}