mlllm.io

# mlllm.io

mlllm.io is a public AI news and builder lab by Sergey Kostenchuk. It publishes short AI news briefs, expanded source-backed articles, public project pages, and practical notes about AI publishing pipelines, agent workflows, and LLM quality gates.

## Key Sections
- https://mlllm.io/start/ - First-reader guide to the site model, feeds, explainers, RSS, and llms.txt.
- https://mlllm.io/news/ - English short-news index for current AI signals.
- https://mlllm.io/ru/news/ - Russian short-news index.
- https://mlllm.io/articles/ - Expanded English articles linked from short news.
- https://mlllm.io/ru/articles/ - Expanded Russian articles linked from short news.
- https://mlllm.io/topics/ - Evergreen topic pages about the site architecture and publishing model.
- https://mlllm.io/projects/ - Public systems and open-source work behind the site.
- https://mlllm.io/blog/ - English builder notes and author essays.
- https://mlllm.io/ru/blog/ - Russian builder notes and author essays.
- https://mlllm.io/rss.xml - RSS 2.0 feed for current English AI news briefs.
- https://mlllm.io/about/ - Author profile and editorial scope.
- https://mlllm.io/community/ - Telegram, Threads, GitHub, and public collaboration links.

## Best Used For
- Building an AI news pipeline -> https://mlllm.io/topics/ai-news-pipeline/
- Understanding TG-NEWS -> https://mlllm.io/projects/tg-news/
- Inspecting a live cursor-trail reveal Agent Skills demo -> https://mlllm.io/projects/mouse-trail-masking-reveal/
- Designing LLM publication QA gates -> https://mlllm.io/topics/llm-publication-qa/
- Structuring one brief and one longform article per story -> https://mlllm.io/topics/story-surface-contract/

## Author Essays
- https://mlllm.io/ru/blog/skill-kotoryy-sozdaet-skilly/ - Скилл, который создаёт скиллы. Авторская статья о Skill Creator V2 как мета-скилле для создания других agent skills: сначала классифицировать работу, определить риск и evidence, сгенерировать skill, проверить, упаковать и оставить будущие skills тестируемыми.
- /ru/blog/glubokaya-taksonomiya-klassov-skillov-dlya-skill-creator-v2/ - Глубокая таксономия классов скиллов для Skill Creator V2. Очищенная публичная companion note к Skill Creator V2. Объясняет, почему плоская таксономия скиллов ломается, какие шесть осей используются, как отличать class, subclass, tag и tool surface, и как таксономия влияет на evidence gates и skill groups.
- https://mlllm.io/blog/the-skill-that-builds-skills/ - The Skill That Builds Skills. An author essay about Skill Creator V2 as a meta-skill for creating other agent skills: classify the work first, define risk and evidence, generate the skill, review it, package it, and keep future skills testable and auditable.
- /blog/deep-skill-class-taxonomy-for-skill-creator-v2/ - Deep Skill Class Taxonomy for Skill Creator V2. A cleaned public companion research note for Skill Creator V2. It explains why flat skill taxonomies break, what six axes are used, when something is a class, subclass, tag, or tool surface, and how the taxonomy affects evidence gates and skill groups.
- https://mlllm.io/ru/blog/kak-mouse-trail-masking-reveal-stal-agent-skill/ - Как Mouse Trail Masking Reveal стал Agent Skill. Авторская статья о Mouse Trail Masking Reveal: связке из двух Agent Skills, которая проверяет совпадение base/reveal изображений и строит canvas-based soft cursor-trail reveal hero с maskCanvas, revealCanvas, настройками эффекта и browser validation.
- https://mlllm.io/blog/how-mouse-trail-masking-reveal-became-an-agent-skill/ - How Mouse Trail Masking Reveal Became an Agent Skill. An author essay about Mouse Trail Masking Reveal: a two-skill Agent Skills workflow that validates base/reveal image alignment and builds a canvas-based soft cursor-trail reveal hero with maskCanvas, revealCanvas, tuning controls, and browser validation.
- https://mlllm.io/ru/blog/kak-rabota-nad-seo-llm-mlllm-io-prevratilas-v-sistemu-skills/ - Как работа над SEO/LLM mlllm.io превратилась в систему skills. Авторская статья о том, как SEO и LLM-friendly оптимизация mlllm.io превратилась в переиспользуемую группировку Codex skills с оркестрацией, валидацией, schema, crawler access, перелинковкой, UX и task-plan процессом.
- https://mlllm.io/blog/how-mlllm-io-seo-llm-work-became-a-skill-cluster/ - How mlllm.io SEO/LLM Work Became a Skill Cluster. An author essay about turning the SEO and LLM-friendly optimization work behind mlllm.io into a reusable Codex skill cluster with orchestration, validation, schema, crawler access, link architecture, UX, and task-plan based execution.
- https://mlllm.io/ru/blog/kak-ya-sozdal-ai-skill-senior-xray-vpn-architect/ - Как я создал AI-скилл Senior Xray VPN Architect. Авторская статья о проектировании AI-скилла Senior Xray VPN Architect для легитимной приватной VPN/proxy-инфраструктуры: роль, safety-границы, режимы работы, reference-файлы, split tunneling, client compatibility, DNS/routing, актуализация источников, линтер и monitoring.
- https://mlllm.io/blog/how-i-built-the-senior-xray-vpn-architect-ai-skill/ - How I Built the Senior Xray VPN Architect AI Skill. An author essay about designing the Senior Xray VPN Architect AI skill for legitimate private VPN/proxy infrastructure: role, safety boundaries, work modes, reference files, split tunneling, client compatibility, DNS/routing, source freshness, linting, and monitoring.

## Do Not Use
- https://mlllm.io/admin/ - private admin area.
- https://mlllm.io/api/ - private/admin and ingestion APIs.
- https://mlllm.io/plans/ - internal planning artifacts.
- https://mlllm.io/server/ - internal server code paths.
- Draft, unpublished, or unreviewed locale URLs.