Complex React components become hard to test and maintain. This skill provides patterns and workflows to refactor high-complexity components by extracting hooks, splitting sub-components, and applying Dify-specific conventions.

Building production-grade LLM applications requires understanding complex architectural patterns. This skill provides proven LangChain patterns for agents, chains, memory management, and tool integration.

Research complex topics efficiently without getting overwhelmed. This skill organizes your research into manageable subtopics and synthesizes findings into comprehensive reports.

Creating AI skills from scratch requires understanding structure, best practices, and validation. This skill provides step-by-step guidance and automated tools to build, scaffold, and validate skills for any domain or workflow.

Finding accurate LangGraph documentation requires navigating multiple pages and filtering through extensive content. This skill automatically fetches the documentation index, identifies the most relevant guides for your question, and retrieves the specific content you need.

Finding academic papers across physics, mathematics, computer science, and quantitative biology can be time-consuming. This skill provides direct access to the arXiv preprint repository to search and retrieve paper titles and abstracts instantly.

Playwright skill automates web testing with visible browser execution. It detects local dev servers, writes test scripts to temporary storage, and handles form filling, screenshots, and responsive design validation without cluttering your project.

Need to implement graceful workflow termination in your AI agent system. This design document provides a Redis-based state management solution with API endpoints and polling patterns for reliable workflow stopping.

FastGPT 工作流支持暂停并等待用户输入。本指南详细说明交互节点的完整开发流程，包括类型定义、执行逻辑、前端组件和国际化支持。

Building AI skills is hard without proper structure. This guide provides proven templates and best practices for creating skills that work reliably and are easy to discover. It includes ready-to-use templates for basic and workflow skills.

Managing Home Assistant configurations, automations, and dashboards requires understanding multiple workflows, tools, and best practices. This skill provides expert-level guidance for deployment workflows, remote CLI access, automation verification, and tablet-optimized Lovelace dashboard development.

Coding agents drift, expand scope, and skip verification — this skill provides a lightweight PM operating loop with templates for briefing, reviewing, and shipping their work. It gives founders and product engineers a repeatable system to delegate, review, and verify agent output without losing control.

Sharing files or sites usually requires setting up hosting, accounts, and configuration. This skill publishes text, files, and entire sites to brewpage.app with a single command, returning a public URL ready to share.

Standard Markdown lacks features for linking notes and organizing knowledge. This skill helps you create properly formatted Obsidian notes with wikilinks, embeds, callouts, and properties.

This skill enables AI assistants to interact with Obsidian vaults via CLI, allowing note creation, search, task management, and plugin development workflows directly from the command line.

Transform your Obsidian vault into a powerful database with table, card, list, and map views. Create dynamic filters, formulas, and summaries without writing code.

Users struggle to create visual canvases in Obsidian without a proper JSON format guide. This skill provides complete reference documentation for creating and editing JSON Canvas files with nodes, edges, groups, and connections.

Reading web pages with WebFetch wastes tokens on ads, navigation, and clutter. Defuddle extracts only the main content as clean markdown, saving up to 80% of tokens when analyzing online articles, documentation, and blog posts.

The ZINC database contains 230M+ purchasable chemical compounds for drug discovery research. This skill provides programmatic access to search by ZINC ID, SMILES, or supplier code, retrieve 3D structures, and perform similarity searches for virtual screening and molecular docking studies.

Working with large datasets that exceed memory limits. Zarr-python enables chunked array storage with compression for efficient cloud-native scientific computing workflows.