5,123 boosters — open source, verified from GitHub, ready to install
Gollem is a Cursor rules booster that establishes development standards and restrictions for Go-based agentic AI applications using MCP, ensuring consistent code quality, English-only documentation, and proper testing practices.
"name": "firecrawl", "description": "Scrape, search, crawl, and map the web with a single command.", "./skills/firecrawl-agent",
Cursor Rules for Starknet documentation that enforces Mintlify technical writing standards and best practices for creating clear, user-centered developer documentation. Ideal for Starknet contributors and documentation maintainers using Cursor.
This is a VS Code color theme extension. There is no runtime code, no build step, and no test framework. The deliverables are JSON theme files and the manifest.
This is the Go implementation of Vibes (main branch). The Python version lives on the branch.
A comprehensive Go testing booster that teaches TDD methodology, table-driven tests, benchmarking, and fuzzing patterns for Claude Code users. Ideal for Go developers aiming to write reliable, well-tested code following idiomatic Go practices.
A Cursor IDE rules file that enforces Python coding standards (PEP 8, Black formatting, Google-style docstrings) for geohash module development. Useful for Python developers using Cursor who want consistent code style enforcement.
A Chinese-language system prompt that transforms simple descriptions into detailed video generation prompts for AI video models, designed for use in Claude, ChatGPT, Cursor, and Windsurf environments.
Universal Skills is an MCP server that enables discovery and installation of reusable skills from GitHub repositories, allowing Claude Desktop and Claude Code users to extend AI capabilities with custom tools and integrations.
Query Google's AI Search mode to retrieve comprehensive, source-grounded answers from across the web. Trigger this skill when the user: 1. Include Current Year (2026) for up-to-date results
Ensure every critical action is logged (vital for UAG/Trust Room).
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…)
"$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "name": "polyakov-claude-skills", "description": "Набор скиллов для Claude: работа с .docx, веб-скрапинг, управление агентами",
Heuristic scoring (no AI key configured).
Analyzes ANY input to find, improve, or create the right skill. Start with least privilege (, , , , ). Only add higher-risk tools when explicitly required:
GPUI is a UI framework which also provides primitives for state and concurrency management. Context types allow interaction with global state, windows, entities, and system services. They are typically passed to functions as the argument named . When a function takes callbacks they come after the p
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。
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。
Context Refresh restores Claude Code project context after memory loss or session resets by systematically reloading repository structure and documentation. Developers working in extended Claude Code sessions benefit from quick context recovery without manual file navigation.
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。
"name": "design-council", "description": "Convene parallel role-specialized peer agents (dynamic roster, plan card first) to debate a cross-domain decision or audit a codebase in real time. Invoking Claude acts as CEO: convenes, routes peer-DMs, arbitrates deadlocks, writes a one-page decision log."
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。
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。
