65 boosters for "learning" — open source, verified from GitHub, ready to install
Carousel Growth Engine autonomously transforms any website into viral TikTok and Instagram carousels, analyzing content, generating images via Gemini, publishing directly to feeds, and optimizing through analytics feedback. Ideal for social media managers, content creators, and marketing teams seeking to automate carousel production at scale.
Corporate Training Designer is an AI agent that helps enterprises design and optimize training programs through needs analysis, instructional design, and effectiveness evaluation. HR leaders, L&D professionals, and training managers use it to create behavior-change-focused curricula and leadership development initiatives.
An AI/ML engineering expert agent that helps developers build, deploy, and integrate machine learning models into production systems with scalable, practical solutions. Ideal for engineers building intelligent features and data pipelines.
Train language models using TRL (Transformer Reinforcement Learning) on fully managed Hugging Face infrastructure. No local GPU setup required—models train on cloud GPUs and results are automatically saved to the Hugging Face Hub. Use this skill when users want to: Use Unsloth () instead of standard
You are an expert at using the TRL (Transformers Reinforcement Learning) library to train and fine-tune large language models. TRL provides CLI commands for post-training foundation models using state-of-the-art techniques: TRL is built on top of Hugging Face Transformers and Accelerate, providing s
Transformers.js enables running state-of-the-art machine learning models directly in JavaScript, both in browsers and Node.js environments, with no server required. Use this skill when you need to: The pipeline API is the easiest way to use models. It groups together preprocessing, model inference,
A skill for fine-tuning and training language models on Hugging Face's cloud GPU infrastructure using TRL, supporting SFT, DPO, GRPO methods and GGUF conversion for local deployment. Developers and ML engineers working with cloud-based model training benefit from this comprehensive guidance.
<!-- AUTO-GENERATED FILE. DO NOT EDIT MANUALLY. --> <!-- Source: tools/ai/llms-base.txt + tools/ai/aicontextheader.md --> <!-- Regenerate with: python tools/ai/generateaicontext_files.py -->
This module focuses on learning the language transition from JavaScript to Python, helping developers quickly master Python programming through comparative teaching. console.log("Hello World"); print("Hello World")
A system prompt that guides LLMs to analyze Factorio game implementations and generate detailed natural language plans for achieving objectives. Useful for developers creating AI-driven game planning systems or educational tools.
"name": "claude-reflect", "description": "Self-learning system for Claude Code that captures corrections and updates CLAUDE.md automatically", "name": "Bayram Annakov",
Frontmatter fields above are primarily for Claude Code / OpenClaw. author: Agents365-ai category: Content Creation
name: sona-learning-optimizer description: SONA-powered self-optimizing agent with LoRA fine-tuning and EWC++ memory preservation type: adaptive-learning
This booster provides expert guidance for developing, debugging, and optimizing Azure AI Document Intelligence applications, covering architecture, security, best practices, and deployment patterns. Developers building document processing solutions on Azure will benefit from its comprehensive troubleshooting and design pattern knowledge.
Procedural memory for AI coding agents. Transforms scattered sessions into persistent, cross-agent memory. Uses a three-layer cognitive architecture that mirrors human expertise development. AI coding agents accumulate valuable knowledge but it's: You've solved auth bugs three times this month acros
Structured approach to improving performance through focused effort, feedback, and continuous refinement. Based on psychologist K. Anders Ericsson's research on expertise acquisition.
用户要动手做 / 研究一个东西，或想把某个已有产出改得更好。这是"重输入、轻输出"短板的解药——逼用户从输入切到输出。 别追求完美，先有一个能跑 / 能看的最小版本。卡在"还没准备好"就是没进改良主义。 针对缺陷提一个改良策略（视为假说，可对可错），动手改，看效果。错了也有用——错误暴露后，下次自动规避这个方向。
用户在纠结"要不要学 X / 学到什么程度 / 精力往哪放"。这是"广度优先、兴趣队列过长"倾向的刹车。 逼问一句：你要解决的具体问题是什么？ 没有具体问题、纯"感觉该学 / 别人都在学"→ 直接进"以后再学"队列，不占当下精力。理解知识的作用，重于知识本身。 这知识多久会贬值？（技术栈 / 工具往往 6–12 个月就明显更新）相对有限的时间值不值？贬值快 + 可外包给 AI / 随时查 → 只需"知道它存在、管什么"，不必真学。
用户学完一个东西想验真伪，或隐约觉得"好像懂了但不踏实"。也是"重输入轻输出"的一次强制输出。 请他用自己的话、把你当外行，把概念讲一遍。别让他背定义——要他解释、打比方。 专挑他含糊带过、用术语糊弄、跳过的环节追问："为什么？""那这个是怎么来的？""举个例子？"命中他答不上来或开始绕的地方。
用户要系统进入一个新领域，或焦虑"学得不够系统 / 不知何时算够"。 用户为什么学 X？（接 的"既定问题"）目的决定图谱画到多细。 从入门点出发、沿父子关系排一条有效路径。颗粒度按需自由切换（领域图 → 细分学科图）。"学到哪算够"= 覆盖到能解决第一步那个目的的节点即可，不必学满。
用户在学 / 接触一个新概念 X（新技术、新算法、新理论、新领域……），尤其觉得"陌生 / 有点难"的时候。难，往往不是智商问题，是它相对用户还存在"没接上的旧知识"。 用一两句话说清 X 到底在干什么——它的核心机制 / 结构是什么。剥掉术语外壳，留下"它本质是一个 "。只有先拿到结构，才能去匹配用户学过的东西。 拿不准就直接问「你学过 吗？」，绝不从正在讲的材料 / 文章作者背景推断用户会什么。
基于 Benjamin Bloom「2 Sigma Problem」研究（1984）的一对一 AI 导师系统。每个课题是一个独立文件夹，通过自适应生成的课程文档 + 用户反馈循环模拟一对一苏格拉底式导师，把学习效果推向 +2σ。学习的主要载体是文档，对话只是辅助确认状态。 所有回复、解释、提问、文档一律使用中文。 触发本 skill 后，以下守则在整个学习交互全程生效——违反字面就是违反精神：
用户说"想学 / 理解 / 搞懂 / 讲讲一个概念 X"时——这是默认入口，一次跑完五视角，用户再选深入哪个。 抓住 X 的本质结构（剥术语），按三猜想给 🎁其实已学过 / 🔗结构同构（字段级对应表）/ 🧩可用已有知识解释，点出元知识。先激发信心，再谈深入。 定位"既定问题"（学 X 解决什么）、现有知识够不够、X 的贬值速度与 ROI，给"够用就停 / 只学最小那块 / 值得深挖"的深度边界。不是劝退，是防止一上来过度钻。
