ljg-skills
A collection of skills for Claude Code that cover analogies, blind spot scanning, book explanations, business model canvases and content cards, with support for both Chinese and English.
4,248 plugins found in public marketplaces and GitHub repositories, indexed and searchable. Plugin skills also install in Codex, Cursor and 4 more agents. In Claude Code, the 99 that get Auto-invokedWhat is this?This plugin ships a FLOW.md router the engine fires, so the matching skill runs itself. No slash command to remember.Learn how → fire as you prompt.
A collection of skills for Claude Code that cover analogies, blind spot scanning, book explanations, business model canvases and content cards, with support for both Chinese and English.
Gives Claude Code users skills for working inside Agent-Native workspace apps, making brand-safe images and videos, editing docs and blogs, comparing UI design directions, and checking how much context a session uses.
A general-purpose AIGC video engine: script to finished film in one pipeline, dramas, ads, product videos, otome games, and more. | 通用 AIGC 视频引擎 , 从剧本到成片一条流水线,漫剧、广告、电商、乙游皆可.
A terminal pair programmer that brings AI help into your coding workflow, with ready-made skills for pull request descriptions, executing structured task plans, and creating new skills. Works with many AI models, so you can keep working where you already work.
Push it to GitHub and send us the link. We read it, then it goes live.
Gives your AI assistant focused skills for reviewing diffs and PRs before shipping, finding the root cause of an error before anyone patches it, auditing agent setup when it misbehaves, running structured research, and fetching or summarizing web pages and PDFs.
A pixel-art office dashboard that shows in real time what each AI assistant is working on, whether it is online, and what it did recently. It supports multiple agents, a Chinese, English and Japanese interface, and can be deployed as a standalone status board.
A Chinese-language writing skill that removes AI writing tells by rewriting only the patterns on an explicit whitelist, leaving all other text and the article's structure untouched. Useful for cleaning up a finished draft.
A design skill that turns a theme, mood, or reference into a sparse paper-texture editorial poster, a ready-to-use image prompt, or a reusable visual system. It is written for Codex, not Claude Code.
Gives a developer one shared set of agent rules and reusable workflow skills, covering tasks such as controlling your signed-in Chrome, redacting and previewing agent transcripts, and checking ClickClack deploys. Use it to keep several local coding agents working the same way.
Manages a shared library of agent skills for Claude Code users, letting you install, update, remove, and deploy skills per agent, organize tags and presets, search, and adopt existing skills.
A local-first Agent Workbench where people and agent teams share one workspace to run multiple agent engines, manage plugins and Skills, and coordinate multi-agent projects and tasks.
Build agents and multi-agent systems with the Swarms framework, covering the Agent class, tools, memory, autonomous loops, and workflows such as sequential, concurrent, and graph-based orchestration.
Gives a machine learning team one place to define features and serve them consistently, so training data and live predictions use the same values. Covers the offline store, online store and feature server.
Plugin for developers of the Civitai codebase, with skills that guide adding new AI model ecosystems, base models, generation support, prompt-enhancement guides, and training support to the app and its trainers.
Moves routing between agents, model selection, guardrails and tracing out of your application code into a separate proxy layer. Agents and models can change without rewriting app code, and tracing covers every agent without extra instrumentation.
Analyzes codebases to find duplicate business logic, check coupling between modules, flag misplaced classes, size components, and plan migrations away from monoliths.
This project implements a Model Context Protocol (MCP) integration between AI agent (Cursor, Claude Code) and Figma, allowing AI agent to communicate with Figma for reading designs and modifying them programmatically.
A skill for agent-to-agent contracts that lets agents transact, negotiate, and pay one another. It gives them a trust framework for doing so without a human in the loop.
Turns plain written content into a polished infographic, so you can skip laying out charts and text by hand. Ships skills that generate infographic syntax from your notes and reports, and render it with templates and themes.
Nuxt UI is a library of accessible, typed Vue components built on Reka UI, Tailwind CSS and Tailwind Variants. It works as a module in Nuxt and as a Vite plugin in plain Vue apps.
Autonomously optimizes Agent Skills the way a model is trained, running repeated experiments and keeping only changes that improve the skill. Inspired by Karpathy's autoresearch.
A search skill for AI agents that runs real-time web searches, vertical domain searches, parallel batch searches, and full-page content extraction. It works anonymously with lower rate limits or with an API key for higher ones.
A template-driven skill for generating data charts, HTML or PDF slide presentations, and full-page HTML reports. It builds charts from the Lupi, Basics, Glance, Maps and Interactive gallery implementations, with a neutral default palette and optional custom colors.
Runs an autonomous loop that modifies code, verifies the result against a mechanical metric, and keeps or discards each change, for iterative improvement, debugging and ship-readiness checks in Claude Code.
Agent skills from Remotion that guide Claude Code and similar AI agents in building videos with Remotion, covering video creation, captions, interactivity and searching the Remotion docs.
A pixel desktop pet that watches Claude Code, Codex, Cursor & other AI coding agents, so you don't have to.
Browser automation for Claude Code users, using persistent named pages through the dev-browser CLI. Use it to navigate websites, fill forms, take screenshots, extract web data, test web apps, and log into sites.
Agent-agnostic skills for AI short-drama production, from a novel to shoot-ready material: character bibles, adaptation outlines, scene & prop bibles, screenplays, opening staging, and storyboards.
Expert Claude Code skills for building flawless n8n workflows using the n8n-mcp MCP server.
Builds an AI Skill that imitates a former partner from imported chat logs, social media screenshots, photos, and your own descriptions, so it speaks in their phrasing and recalls shared places. Meant for personal memories and emotional processing only.
Generates type-safe TypeScript API clients, query hooks, Zod schemas and MSW mocks from OpenAPI or Swagger specs in YAML or JSON. Built for developers wiring frontends and backends to REST APIs.
Skills for Genkit, a framework for building agentic AI apps in JavaScript, Go, Dart, and Python. Covers TypeScript conventions and guidance for writing Genkit CLI commands and MCP server tools.
An MCP server and CLI that gives AI agents tools to build, test, run, debug, and automate UI on iOS, macOS, watchOS, tvOS, and visionOS projects. Includes skills for both the MCP server and the CLI.
A control plane for long-horizon agents that keeps work moving across sessions, with durable state, Goal and Todo tracking, delegation, benchmark analysis and change-quality checks for Claude Code users.
Provides policy enforcement, zero-trust identity, execution sandboxing, audit trails, and reliability engineering for autonomous AI agents, with commands and MCP servers for Claude Code users.
Duplicate code detector for 220+ languages, plus dead code, complexity hotspots, duplication and health trends over git history, and one health score for the whole codebase.
A SwiftUI review skill that checks code for modern API use, deprecated APIs, maintainability, performance, and accessibility such as VoiceOver. Useful when reading, writing, or reviewing SwiftUI projects.
Adds Preline UI components and blocks to HTML files through the Preline MCP server, and generates, previews, and validates Preline theme CSS. Built on Tailwind CSS.
A desktop client for running AI agents across multiple model providers, extendable with MCP servers and skills, and controllable from a phone. Built with Electron and Next.js, with Feishu/Lark tools for documents, bitables, calendars and messaging.
Converts content from WeChat articles, web pages, YouTube, podcasts, PDFs and Markdown into NotebookLM, then generates podcasts, slides, mind maps or quizzes from it. Supports a deep analysis mode and automatic Feishu document creation.
Converts content from sources like web pages, WeChat articles, YouTube, podcasts, PDFs and Markdown into NotebookLM and generates podcasts, slides, mind maps or quizzes from it.
A terminal window manager for Claude Code users, with tiling panes, workspaces, and sessions that survive restarts. The skill lets Claude read and write other panes, run work and wait on it, and report its own state.
A self-hosted dashboard for operating AI agents, letting you dispatch tasks, inspect runs, review failures, track spend, and coordinate agent runtimes from one local interface backed by SQLite. The included skill helps Claude Code register agents, manage tasks, and query status through its APIs.
Skills for building virtualized React lists, grids, and tables that handle variable-size items and sticky headers, plus chat message lists, data tables, and reactive state management with the Virtuoso packages.
Rewrites Korean text that was drafted by an AI so it reads like human writing. It changes only style, rhythm and phrasing, leaving facts, claims and figures untouched. Includes a quick scan and a commit-message variant.
Skills for installing, configuring and operating the vLLM Semantic Router on Docker or Kubernetes, changing its routing contracts, calibrating routing recipes against live model backends, and handling maintainer issue and release work.
A Claude Code skill that rewrites dense, ambiguous English into ASD-STE100 Simplified Technical English (STE), the controlled-language standard the aerospace and defense industry built so aircraft maintenance instructions cannot be misread.
Fast and Accurate Code Search for Agents. Uses 99% fewer tokens than grep+read.
Provides skills for DeepChat, an AI assistant that can drive desktop apps, work with Feishu/Lark documents and spreadsheets, and save and recall approved memories across conversations. Useful for automating tasks in real applications and keeping context between sessions.
Turns a rough feature idea into a written spec through guided questions, then has the agent build it and checks the result against that spec. Instead of hand-checking whether the work is finished, you get an automatic evaluation gate that runs in your coding agent.
Skills for building Atomic Agents, a Python framework for AI agents: agents with typed input and output schemas, tools, and context providers that inject dynamic blocks into the system prompt. Also includes a release skill for publishing new versions.
Runs a math modeling contest from start to finish: it analyzes the problem, builds and solves the model, produces the code and charts, and drafts the paper in Typst or LaTeX, so you get a complete submission-ready paper instead of assembling each stage by hand.
Lets AI agents drive your own Chrome, with your existing logins, to read pages, fill in forms and pull information from sites like Twitter, Reddit and GitHub. Works with OpenClaw's browser directly, so no extra extension is needed.
Runs multiple coding agents in parallel, each in its own git worktree, with support for remote SSH, pull request review and issue ingestion. Works with any provider.
The CLI and skills that turn any coding assistant into an expert at creating, evaluating, and deploying AI agents on Google Cloud.
Skills for researching social media, covering ad library teardowns, audience analysis, comment mining, competitor strategy, and repurposing public content into new posts and scripts.
Gives Claude Code skills for adding CSS transitions and animations to web apps, covering badges, modals, dropdowns, page transitions and motion polish. Includes a skill that animates UI as if an AI agent were generating it.
An open-source desktop AI agent for data analysis, slide and document creation, video and web research, and scheduled tasks. It runs tools on your computer and accepts commands from your phone through WeChat, Feishu, DingTalk, or Telegram.
Breaks a complex task into steps and hands each one to a specialized agent for research, coding, design or unattended automation. It also carries memory across sessions so the working setup can improve itself over time.
Claude Code skills for generating or editing images with GPT Image through a packaged CLI and reference gallery, plus reverse-engineering image prompts from reference photos.
Webwright is now CUAWright. Alongside browser automation, you can now run desktop tasks in an Ubuntu VM. Use cuawright web for browser tasks and cuawright desktop for desktop tasks. Existing webwright commands and Python imports still work.
Automates Excel data analysis end to end: reads multi-sheet workbooks, cleans the data, filters and aggregates it, flags outliers and threshold breaches with highlighting, and exports the results as Excel or CSV. Large files are converted to Parquet for faster reading.
Opinionated Oxlint lint rules that flag low-evidence and low-signal TypeScript and JavaScript patterns in your code. You copy the rules into your repository and adjust them to your team's standards, and a bundled skill handles the initial setup and later upgrades.
Sets up Claude Code and Codex for a project with minimal configuration, and adds skills for pull requests, releases, documentation updates and BMad-Method initialization.
A research assistant plugin for Claude Code that supports literature review, paper writing, citation verification, Zotero and Obsidian integration, and code review and debugging for academic and software work.
Gives Claude Code guidance on Anthony Fu's preferred JavaScript and TypeScript tooling, conventions and project setup, plus Nuxt, Nitro and Pinia usage. It also drafts GitHub pull requests with clear titles and bodies, so you spend less time on setup decisions and PR write-ups.
A Claude Code plugin for designing dashboards, admin panels, SaaS apps, and settings pages with consistent, intentional interfaces. It remembers design decisions across sessions and supports reviewing and refining existing UI, but is not meant for marketing sites.
Claude Code skills for LinkedIn that draft posts, comments and replies in your voice, plan a week of content, extract hook formulas from viral posts, and analyze post engagers. Nothing is published without your approval.
Runs recon on a target, tests it for real, reportable vulnerabilities, checks each finding against a strict validation gate, and writes a report ready to submit to bug bounty platforms. It also remembers patterns across targets and resumes unfinished sessions.
A Claude Code plugin for bug bounty hunters, bundling skills for reconnaissance, vulnerability hunting such as IDOR, SSRF and XSS, black-box audits of deployed AI agents, and CI/CD pipeline security.
A Claude Code plugin with skills for building durable backend AI agents with the eve framework, plus skills to investigate issue reports, look up agent runs, and run security and correctness audits of code diffs.
Skills for analyzing and comparing Xcode builds, test runs, app bundles and cache runs with Tuist, including selective testing and cache invalidation diagnosis. Useful for iOS developers tracking build regressions.
Gives AI agents a way to connect to outside tools and services through MCP servers, so they can act on your accounts and data without you wiring each integration by hand. Also includes a documentation lookup skill that answers questions about libraries and frameworks with code examples.
An MCP server that lets Claude create and open diagrams, flowcharts, and mockups in the draw.io editor.
A Claude Code skill that turns any codebase into a beautiful, interactive single-page HTML course. Point it at a repo.
A context graph that indexes codebases and SDLC knowledge so Claude Code can search code and keep project memory. Includes skills for backfilling newly added repos and composing graph mutation plans.
Local, clinical AI skills for healthcare data, covering de-identification of protected health information, clinical entity extraction, FHIR bundle assembly, variant annotation, and audit trails for compliance review.
Gives coding agents ready-made help for everyday C# work, such as running one-file C# apps, calling native libraries from .NET, setting up keyless NuGet publishing, and writing fast SIMD code.
Chinese localization of Matt Pocock's skills for Claude Code, covering code review against standards and specs, bug diagnosis, domain modeling, and codebase design. Includes an ask-matt router for choosing the right skill.
An open-source tool to remove Gemini watermarks from AI-generated images with high-fidelity, reproducible results on supported outputs.
A self-hostable personal AI assistant that connects any model provider, builds a knowledge base from your own documents, and lets agents browse the web, run code and call MCP tools for you. Its skills cover notes, reminders, secrets and feed monitoring on macOS.
Generates production-ready Lottie animations from a text prompt, so you can create logo, icon, loader and UI motion without hand-writing the JSON. Previews them in a local Skia player while you work with Claude Code, Codex or another coding agent.
Languages: English | 繁體中文 | 简体中文 | 日本語 | 한국어 0.4 launch teaser: every frame is real ASF output.
Skills for building iterated agentic coding loops, designing control loops for a codebase, improving CLAUDE.md instruction files, narrowing React prop types, and running a structured research-plan-implement workflow.
Audits and rewrites text to remove AI writing patterns, often called AI-isms, from any content. Supports detect-only scans, in-place file edits, and voice profiles to keep the author's tone.
Plugin for github/gh-aw, GitHub Agentic Workflows, which lets you write repository automation as markdown workflows that run AI agents in sandboxed GitHub Actions. It includes a Playwright skill for browser automation and web page testing.
Install language servers and their toolchains so the lsp tool works, and give recipes for finding symbol definitions, usages, types and compiler errors in Go, TypeScript and other languages. Built for Claude Code users who want code navigation.
Install language servers and toolchains for Go, TypeScript, Python and Rust, and guide the lsp tool to find symbol definitions, usages, types and compiler errors.
Skills for NVIDIA Model Optimizer, a library for compressing and speeding up machine learning models through quantization, pruning, distillation, and neural architecture search. Includes skills for benchmarking model kernels and comparing evaluation results in MLflow.
Gives your AI coding agent local-first web search, fetching, crawling, page diffing, and structured data extraction over MCP. Pages already fetched are cached locally and reused without new web requests.
A self-contained bundle of Claude skills and slash commands for bug bounty hunting and external red-team work, covering recon, Android APK analysis, and vulnerability reporting for Bugcrowd and HackerOne.
The open-source AI research workbench for scientific research and agent workflows. Local-first, model-agnostic desktop app with extensible skills, MCP tools and connectors, Python/R execution and traceable artifacts for reproducible research on macOS, Windows and Linux.
Lets an AI assistant work on your own local code, reading, editing, searching and running commands in your real projects, through a secure connection you control. Nothing is uploaded to a third party, and you approve each connection with your own password.
Skills for the bb agentic IDE: configure the Account Pooler and quota failover, manage recurring automations, annotate browser tabs, sign in to a getbb.app account, and control the hosted cloud AI features.
A visual canvas for designing AI agent workflows that exports them as Markdown for Claude Code. Also includes skills for Jira-driven planning and autonomous development loops.
The programming language for agents. Zerolang is an experimental graph-native programming language where the semantic graph is the program database. Humans ask for outcomes. Agents query the graph, submit checked edits, and prove the result.
Adds the Agentation visual feedback toolbar to a Next.js project so you can annotate page elements for an AI agent. A self-driving skill lets Claude critique a page and add design annotations on its own.
Lets AI coding agents drive and verify mobile apps on iOS, Android and Amazon Vega TV simulators or devices. Agents can launch apps, inspect the live UI, tap, type and check their own code changes, plus run exploratory bug hunts.
Gives Claude Code a persistent second brain stored as plain markdown in an Obsidian vault. Notes rewrite themselves as new sources arrive, and scheduled agents handle research synthesis and note capture.
A skill that turns natural language descriptions of workflows, architectures, or concepts into Excalidraw diagram JSON files. Useful for coding agents that support skills.
Gives Claude Code access to Exa for web search and crawling, plus multi-step web research, list-building, and structured output. Useful when one search is not enough.
A skill that scaffolds, extends, and verifies reactive local-first JavaScript or TypeScript apps built on TinyBase, a small in-memory store for tabular and key-value state with sync support.
Skills for building native-feeling mobile app screens in Expo and React Native, based on patterns from real app screens and flows. Includes a skill for researching those screens through the Appllama MCP before building.
Plays game character voice lines and shows on-screen banners when your AI coding agent finishes or needs permission, so you can tab away without missing it. Works across many coding agents and IDEs.
Extracts the full design system from any website URL, producing design tokens, Tailwind, Figma variables and shadcn/ui themes, plus a skill that styles UI from the extracted tokens.
Lets Claude Code users query databases in plain English, convert questions into SQL, design schemas and tables, and work with WhoDB hosts and projects through MCP tools.
Gives Hermes users a governing layer that frames each request, picks the workflow and evidence checks, and records honestly what happened. It covers planning, research, coding handoffs, operations and project memory.
Skills for running empirical economics analysis end to end in Python, Stata, or R, covering DID, RD, IV, synthetic control, and double machine learning. Each skill follows the conventions of applied economics papers and produces paper-style results.
Generates, polishes, reviews, amends and publishes command-line tools for AI agents, so they can run with fewer wasted steps and tokens. Includes skills for importing and publishing CLIs from a shared library.
Writes complete Chinese novels from scratch in chapters, with suspense hooks at chapter ends, memory of creative preferences, and automatic checks, for long web serial fiction. Open source and free.
Keeps long OpenClaw conversations from losing earlier messages. Instead of dropping old context when the window fills, it summarizes it in layers, so you can still recall details from far back while staying within the model's token limit.
A vision bridge that lets text-only coding models such as DeepSeek and GLM read images. When an image path or URL appears in the conversation, it returns structured JSON with OCR text, layout, and semantic content.
A skill for building premium scroll-driven landing pages for service, product, food, and drink brands. It helps plan the visitor journey and page layout, then build animated heroes and bespoke scroll effects.
Gives Claude Cowork and Claude Code users a shared place to pick up extra abilities for jobs like simple topic explainers, interactive plans, AI media generation and Playwright test audits. Each one is a ready-made add-on, so you skip building the tool yourself.
Industrial IoT platform for connecting devices over Modbus, OPC UA and MQTT, with device management and time-series data. Includes an fde-ontology skill that runs an automated pipeline from multi-source input to code generation.
Lets you query and synthesize content from your own compiled document knowledge base, with skills to generate HTML slide decks and review them. Works without a vector database and handles long, multimodal documents.
Runs structured deliberation across several reasoning perspectives, so a user can weigh a hard decision from multiple angles before settling on one path.
A WYSIWYG markdown editor so you can edit markdown files like a Google Doc or Notion page, with a desktop app and web UI for navigation, search, and wiki-style linking. It also ships skills for working in codebase wiki, personal CRM, and knowledge base projects.
Powerful, elegant and flexible test framework for Kotlin with assertions, property testing and data driven tests.
© 2026 Flowy
Built for Claude Code · Not affiliated with Anthropic
Logos are trademarks of their owners.