skilld
All skills
yzlnew avatar

yzlnew/infra-skills

mainGitHub
README badge for yzlnew/infra-skills

Repository statistics

  • Indexed skills

    8

  • Skill groups

    1

  • GitHub stars

    0

  • Forks

    0

Skills

8 total

/openai-dotcom-viz

>- Build figures in OpenAI's blog / research / system-card "dotcom" visual style — both (a) bar charts (monochrome bars with a darker same-hue stroke, rounded corners, a black y-axis with outward ticks and no gridlines, circle legend markers left-aligned to the y-axis, value labels above bars, angled category labels) and (b) flow / process diagrams (rounded boxes, monospace uppercase pills, pink highlights, thin open-chevron "arrow" connectors with rounded corners, dashed = negative branch). Everything uses the real OpenAI Sans font and is emitted as self-contained HTML/SVG with zero dependencies. Use this whenever the user wants a chart, pipeline/flow diagram, or figure that looks like OpenAI's website / posts / papers, asks to reproduce an OpenAI figure, or is building a deck or page in an OpenAI-flavored style. Prefer this over generic charting (matplotlib/Chart.js/Recharts) or generic diagram tools (Mermaid) when the aesthetic should match OpenAI.

/hf-architecture-tikz

Draw Sebastian-Raschka-gallery-style TikZ architecture diagrams for any HuggingFace decoder-only LLM, with per-block parameter formulas and concrete numbers. Supports MHA, GQA, MLA, DeepSeek-V4-Flash (Hyper-Connections + Sparse Attention with learned indexer), dense and MoE FFNs (incl. hash routing), and MTP heads. Use when the user asks to visualize / diagram / illustrate a transformer or LLM architecture (DeepSeek, Qwen, Llama, Mistral, gpt-oss, etc.), wants a Raschka-style figure, or wants a TikZ/LaTeX rendering of an HF model.

/slime-user

Guide for using SLIME (LLM post-training framework for RL Scaling). Use when working with SLIME for reinforcement learning training of language models, including setup, configuration, training execution, multi-turn interactions, custom reward models, tool calling scenarios, or troubleshooting SLIME workflows. Covers GRPO, GSPO, PPO, Reinforce++, multi-agent RL, VLM training, FSDP/Megatron backends, SGLang integration, dynamic sampling, and custom generation functions.

/tilelang-developer

Write, optimize, and debug high-performance AI compute kernels using TileLang (a Python DSL for GPU programming). Use when the user requests: (1) Writing custom GPU kernels for AI workloads (GEMM, Attention, MLA, etc.), (2) Optimizing existing TileLang code for NVIDIA, AMD, or Ascend hardware, (3) Implementing non-standard operators (like DeepSeek MLA, FlashAttention variants), (4) Debugging TileLang compilation or runtime errors, or (5) Cross-platform kernel development targeting multiple GPU vendors.

GitHub repository metadata loaded, but the SKILL.md file scan is unavailable right now.

Add to your README

The badge links readers to this page. It shows the skilld mark and no counts, and it follows the reader's light or dark GitHub theme.

<a href="https://skilld.dev/gh/yzlnew/infra-skills"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://skilld.dev/b/yzlnew/infra-skills?theme=dark"> <source media="(prefers-color-scheme: light)" srcset="https://skilld.dev/b/yzlnew/infra-skills?theme=light"> <img alt="Skill repository on skilld.dev" src="https://skilld.dev/b/yzlnew/infra-skills?theme=light"> </picture> </a>