All authors
Claude Skills by michalstrnadel
github.com/michalstrnadel8 skills0 installs8 views
- Source To SkillTurn a YouTube video, playlist or channel, podcast or audio/video file (local Whisper), academic paper (PDF / arXiv URL), book (EPUB, or PDF book via --type book), web article, or GitHub repository - or several of them at once - into an installable agent skill, and refresh such skills later. Use when the user runs /source-to-skill, or asks to convert a video, talk, lecture, podcast, episode, recording, playlist, course, channel, paper, PDF, book, EPUB, article, blogpost, or GitHub repo into a...Votes: 0GitHub stars: 5
- Attention Is All You NeedThe Transformer paper (Vaswani et al., NIPS 2017, arXiv 1706.03762) distilled - scaled dot-product and multi-head attention, encoder-decoder stacks, sinusoidal positional encoding, the warmup learning-rate schedule, base/big hyperparameters, WMT 2014 translation and constituency-parsing results, and the ablation table. Load when explaining, implementing, or checking details of the original Transformer, comparing self-attention against recurrent or convolutional layers, or citing the paper's n...Votes: 0GitHub stars: 5
- Building Effective AgentsAnthropic's guide "Building Effective AI Agents" (Dec 2024) - the workflow vs. agent distinction, the augmented LLM building block, five workflow patterns (prompt chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer), when an autonomous agent is justified, and how to design agent tools (ACI). Load when designing or reviewing an LLM application or agent, choosing between a single call, a workflow and an agent, picking a framework, or writing tool definitions.Votes: 0GitHub stars: 5
- FlaskWorking reference for Flask (pallets/flask), the Python WSGI micro web framework built on Werkzeug, Jinja and Click, distilled from the official docs in the repository. Covers routing and views, the app/request context and request lifecycle, application factories and blueprints, configuration (every built-in config key with its default), the flask CLI and dotenv, Jinja templating, error handling, logging and debugging, testing with pytest, production deployment (Gunicorn, Waitress, mod_wsgi, ...Votes: 0GitHub stars: 5
- How Transformers WorkHow the Transformer and its attention mechanism work, synthesized from the original paper (Vaswani et al., "Attention Is All You Need"), 3Blue1Brown's "Attention in transformers, step-by-step" video, and Jay Alammar's "The Illustrated Transformer". Load when explaining or implementing self-attention, Q/K/V, multi-head attention, masking, positional encoding, encoder-decoder vs decoder-only, transformer dimensions and parameter counts, or when reconciling the different notations and intuitions...Votes: 0GitHub stars: 5
- Intro To Large Language ModelsAndrej Karpathy's 1-hour "Intro to Large Language Models" talk (Nov 2023) distilled - what an LLM is (two files - weights plus run code), how it is trained (pretraining as lossy compression of the internet, fine-tuning into an assistant, RLHF), scaling laws, tool use and multimodality, System 1/2 thinking, self-improvement, the "LLM OS" analogy, and LLM security (jailbreaks, prompt injection, data poisoning). Load when explaining how LLMs work to a non-specialist, reasoning about training sta...Votes: 0GitHub stars: 5
- Neural Networks 3blue1brownCourse skill distilled from 3Blue1Brown's "Neural networks" YouTube playlist (10 videos) - from a plain MNIST digit classifier (neurons, weights, biases, sigmoid/ReLU), through cost functions, gradient descent, stochastic gradient descent and backpropagation with its chain-rule calculus, to large language models and transformers (tokens, embeddings, softmax and temperature, attention heads with query/key/value, masking, MLP blocks storing facts, superposition, the full GPT-3 175B parameter ta...Votes: 0GitHub stars: 5
- Pro GitGit mental models, commands and workflows distilled from Pro Git (2nd ed., Chacon & Straub). Load when explaining how Git works (snapshots, the three trees, branches as pointers, objects and refs), choosing between merge and rebase or a team workflow, undoing or rewriting history safely (reset, revert, rebase -i, reflog recovery), debugging with bisect/blame, setting up a Git server, hooks or .gitattributes, working with GitHub PRs, submodules, or bridging Git to SVN, Mercurial or Perforce.Votes: 0GitHub stars: 5