9 boosters for "txt" — open source, verified from GitHub, ready to install
Rules and patterns for ML demos on Hugging Face Spaces with ZeroGPU hardware. Covers , duration and quota tuning, process isolation, the CUDA availability model, concurrency safety, and CUDA build constraints. This skill is for Gradio SDK Spaces using ZeroGPU hardware. Docker and Static Spaces canno
1. Fetch homepage HTML (curl or WebFetch) 2. Detect business type (SaaS, Local, E-commerce, Publisher, Agency, Other) 3. Extract key pages from sitemap.xml or internal links (up to 50 pages)
This Skill provides Claude Code with specific guidance on how to adhere to coding standards as they relate to how it should handle global conventions. For details, refer to the information provided in this file: global conventions
幫使用者把想法變成高品質的 AI 生成內容 (圖片、影片、音樂)，核心工作是 寫對每個平台的 prompt 以及 必要時自動操作網站。 1. Intent Parser — 一句話拆 9 slot (媒體/長度/畫面比/主題/風格/角色/場景/音訊/語言) 2. Fill Defaults — 沒講的用預設 (video 預設 10s 16:9；動漫預設 Shinkai+Ghibli；電影預設 Deakins DP…)
DyNote 是面向 Codex 这类 Agent 的抖音学习工具，不是一次性摘要器。核心原则是“数据资产先行，学习笔记后置”：先把字幕/转写、评论、元数据和 AI 快读沉淀为可复用资产，再按用户需求生成可追溯的学习笔记、总结和写作材料。默认目标不是字幕工程文件，而是先落一份原始数据包：、、、、、、，并用 归档可复用资产。默认把独立字幕轨或本地自动语音识别转写当作事实主干；当转写密度低、任务需要画面理解或用户只要快速筛选时，再用已登录抖音网页版的“问AI / 识别画面”补充，豆包只作为抖音 AI 不可用时的备用快读或待核验假设。 联网或登录态操作必须先使用 。不要读取、复制或打印 Cooki
当用户请求涉及以下任一主题时使用本 skill： 使用本 skill 处理任务时，优先输出：
Turn a folder of raw files into a Markdown vault that an LLM can grep, and then answer questions over that vault responsibly. source file, carrying retrieval frontmatter (abstract / tags / synonyms) + a
An MCP server that integrates llmstxt documentation fetching into Claude Desktop and Claude Code, enabling developers to access and reference documentation within their AI-assisted workflows.
A system prompt for financial analysis and workflow orchestration designed to enable AI assistants (Claude, ChatGPT) to function as institutional finance engines with structured reasoning and tool integration capabilities.
