Skills

Official VueUse agent skills with progressive disclosure design, sending function overviews first then loading detailed usage on demand for minimal token consumption.

Naval Ravikant 视角 role-play skill that responds in first person as Naval, grounded in his books, podcasts, tweets, and decisions. Codifies five mental models (leverage, specific knowledge, desire-as-contract, redefine-the-word, pain → systemic solution), eight decision heuristics (permissionless leverage, calendar test, indecision = no, manual test, partisan test, desire audit, transform pain into system, behaviour over words), and an explicit expression DNA, plus an Agentic protocol that uses WebSearch for anything fact-dependent before answering.

from github/awesome-copilot

Look up K League 1/2 match results and live standings by date and team via the official K League JSON endpoints and the `kleague-results` npm package — no HTML scraping, supports team name or code filters like `FC서울` / `K09`.

A Claude Code plugin marketplace for modern Swift/SwiftUI development with Swift 6.2, The Composable Architecture (TCA), and professional iOS/macOS features.

.NET 10 and C# 14 best practices skill covering minimal APIs, modular monolith patterns, feature folders, HTTP resilience, Options pattern, Channels, and validation

Apple Reminders management skill using the remindctl CLI on macOS for listing, adding, editing, completing, and deleting reminders with list filtering, date support, and JSON/plain output

Full-stack Longbridge developer skill spanning the `longbridge` CLI (quotes, K-line, positions, portfolio, news, insider trades, statements), Python and Rust SDKs (QuoteContext / TradeContext), MCP server, and llms.txt docs integration across HK/US/CN (SH/SZ)/SG/Crypto markets. Defines an investment-analysis workflow that pulls live data and Longbridge-first news (only falling back to WebSearch when needed) and uses the `<CODE>.<MARKET>` symbol convention everywhere.

Retrieves and provides authoritative, up-to-date OpenAI developer documentation with precise citations for developers seeking guidance on APIs, SDKs, and product capabilities.

A collection of 65 Claude Code skills and 9 workflows with progressive disclosure, covering full-stack development including context engineering and Atlassian integration.

from lijigang/ljg-skills

A skill for downloading high-quality audio from YouTube, SoundCloud, Spotify, and other platforms using yt-dlp and spotdl with playlist support and metadata embedding.

Agent-powered lead intelligence and outreach pipeline that replaces Apollo, Clay, and ZoomInfo by doing signal scoring, mutual ranking, warm-path discovery, voice modeling from real sources, and channel-specific outreach across email, LinkedIn, and X. Requires Exa MCP plus X API, with optional LinkedIn, Apollo/Clay, GitHub MCP, and Mail.app integrations.

Retrieves and updates Jira tickets from your AI coding workflow via the `mcp-atlassian` MCP server (tools like `jira_search`, `jira_get_issue`, `jira_transition_issue`, `jira_add_comment`) or direct REST API v3 `curl` calls. Includes templates for extracting testable acceptance criteria and posting structured progress comments tied to branches and PRs.

One-shot publisher that converts PPTX, PDF, HTML, or Google Slides into a live GitHub Pages URL — handling repo creation, file conversion, Pages enablement, and large-file external-asset mode automatically. Uses the gh CLI plus python3 (and optional poppler-utils) under the hood.

Generates ArchiMate enterprise architecture diagrams using PlantUML's `<archimate/Archimate>` stdlib macros — Business/Application/Technology/Motivation/Strategy/Implementation layer elements, all `Rel_*` relationship macros with `_Up/_Down/_Left/_Right` directional suffixes, and `rectangle "Layer" { ... }` grouping. Best for TOGAF viewpoints, layered EA modeling, motivation analysis, capability maps, and migration planning, always wrapped in a ` ```plantuml ` fence.

Iteratively improves pull requests by addressing CI failures and reviewer feedback until the PR is ready for merging.

Nuxt SEO meta-module covering robots.txt, sitemap.xml generation, dynamic OG image creation, JSON-LD structured data (schema.org), breadcrumbs, and canonical URL configuration.

Command line tools for the ATXP ecosystem to create projects and run demos for building web agents with OAuth and payment capabilities.

Official Nx AI agent configuration artifacts providing skills, subagents, and commands that enhance AI coding assistants with Nx-specific capabilities across Claude Code, Cursor, Copilot, and Gemini.

AI agent skills designed for solopreneurs (One Person Company), providing specialized tools for business operations, development workflows, and productivity automation through Claude Code.

Official Apollo GraphQL agent skills for AI coding agents, covering Apollo Client, Apollo Server, Apollo Connectors, and GraphQL development workflows.

A product management skill that guides AI agents through developing Product Requirements Documents (PRDs) using battle-tested frameworks.

Guides mobile-first responsive design with Tailwind CSS, covering progressive enhancement from small screens to larger breakpoints.

A GitHub Copilot skill from the awesome-copilot collection that helps AI agents write, analyze, and run CodeQL queries for automated security vulnerability detection.

Official Sentry agent skills for employees, providing code review guidelines, commit conventions, PR creation/iteration workflows, bug finding, Claude settings auditing, and CI integration following the Agent Skills open format.

Critical reasoning skill that applies structured challenge methods across 5 modes to stress-test ideas, plans, and decisions, acting as a devil's advocate, pre-mortem analyst, or red team reviewer.

Agent skills for Vue 3 development, derived from real-world issues and providing specialized patterns for AI agents to enhance Vue development workflows.

Vue 3 Options API best practices covering TypeScript integration, method binding, lifecycle hooks, computed properties, and common gotchas like arrow function pitfalls and stateful method lifecycle management.

Multi-agent harness inspired by Anthropic's March 2026 harness-design paper that separates Generator from Evaluator in an adversarial GAN-style loop to drive quality far beyond single-agent self-critique. Use for application builds, frontend design, or full-stack work where AI-slop aesthetics are unacceptable and you want to invest $50–200 on production-quality output.