Skills

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

Orchestrates durable, resumable workflows that survive restarts, pause for events, retry on failures, and coordinate multi-step operations.

A comprehensive performance optimization system for AI agent harnesses, providing production-ready skills, hooks, rules, MCP configurations, and development patterns evolved over months of intensive daily use across Claude Code, Codex, Cursor, and other AI coding tools.

Incident response skill from Anthropic Knowledge Work Plugins for managing security incidents with structured triage, investigation, and remediation workflows.

Best practices skill for Node.js development by Matteo Collina. Part of a curated collection of AI-assisted development skills that includes expertise in Fastify, TypeScript, OAuth, Git/GitHub workflows, and more.

Comprehensive development guidelines for Payload CMS projects, covering collections, fields, hooks, access control, queries, database adapters, and plugin development with TypeScript-first patterns.

A Claude Code plugin marketplace from Trail of Bits providing skills for AI-assisted security analysis, testing, and development workflows.

A Starchild skill that teaches AI agents how the browser Preview panel works, including reverse proxy architecture, iframe-based preview tabs, relative path requirements, and troubleshooting blank or broken preview screens.

Identifies potential bugs and security vulnerabilities in code changes across a branch, providing targeted insights for developers.

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.

from lijigang/ljg-skills

Part of Anthropic's knowledge-work-plugins collection that turns Claude into a role-based specialist. The debug plugin is one of 11 open-source plugins covering productivity, sales, customer support, product management, marketing, legal, finance, data, enterprise search, and bio-research, each bundling skills, connectors, and slash commands for specific job functions.

Runs four-phase automated browser QA on a deployed URL via a browser MCP (claude-in-chrome, Playwright, or Puppeteer): smoke test (console errors, network, Core Web Vitals), interaction test (nav links, form valid/invalid states, auth flow, critical journeys), visual regression at 375/768/1440px with dark mode, and accessibility (axe-core, keyboard nav, screen reader landmarks). Outputs a structured QA report with a SHIP/SHIP WITH FIXES/BLOCK verdict.

A browser-based tool for reviewing AI agent plans and code diffs with inline annotations, structured feedback, and encrypted sharing links, compatible with Claude Code, Copilot CLI, Gemini CLI, and more.

Provides 5 decision frameworks for workload-aware ClickHouse architecture decisions, covering ingestion strategy, join patterns, late-arriving data, time-series partitioning, and real-time pre-aggregation for production deployments.

Retrieves and downloads Douyin videos with support for short and full URLs, extracting video details like title, author, and stats.

OKX OnchainOS skill that signs x402 payment authorizations via a Trusted Execution Environment (TEE) so AI agents can transact with payment-gated web resources — part of the broader OnchainOS skills suite covering wallets, DEX swaps, DeFi, and on-chain gateways across Solana, Ethereum, Base, BSC, and 20+ chains.

crewAI architecture guide that mandates scaffolding with `crewai create flow <project_name>` (underscores, never hyphens) and then choosing the simplest abstraction — `LLM.call()`, `Agent.kickoff()`, `Crew.kickoff()`, or `Flow` — for the job, defaulting to Flow for any production app so state, routing, and error handling are available. Covers `@CrewBase` wiring of `agents.yaml` / `tasks.yaml`, `{variable}` interpolation via `crew.kickoff(inputs={...})`, and Pydantic `response_format` for structured output.

Skill for researching top-performing Xiaohongshu posts and writing optimized notes with title, body, tags, and images, with optional publishing via Xiaohongshu MCP

A Starchild official skill for creating charts and data visualizations within AI agent conversations, installable via the skills CLI for use with compatible coding agents.

A skill from ECC (Everything Claude Code) that distills and optimizes agent rules for agentic workflows. Part of the harness-native operator system featuring skills, instincts, memory optimization, and security scanning across multiple AI agent harnesses.

A collection of 136+ practical Claude skills for office tasks with a 39-tool Office MCP Server covering PDF (extract, merge, split, OCR), spreadsheets, documents, conversions, and presentations, plus an extensible knowledge base and pre-built AI agent personas.

Guides iOS app design by evaluating and optimizing interfaces against Apple's Human Interface Guidelines, ensuring native, intuitive, and platform-consistent user experiences.

Creates world-class PowerPoint presentations with professional design quality inspired by Apple keynotes, Microsoft product launches, and Google I/O. Features multiple brand styles, sophisticated animations, consistent visual hierarchy, and 2024-2025 presentation design trends.

Design, implement, and refactor Ports & Adapters (hexagonal) systems across TypeScript, Java, Kotlin, and Go — with a domain model, use-case application layer, inbound/outbound ports, and adapters wired at a single composition root. Use when introducing domain boundaries, refactoring framework-heavy code, supporting multiple transports for one use case, or swapping infrastructure.

Translates Figma designs into production-ready code with pixel-perfect accuracy using design context, screenshots, and project conventions.

Captures desktop or system screenshots with flexible save locations and cross-platform support for full screen, specific apps, or pixel regions.

Helps create effective AI image generation prompts for DALL-E 3, Midjourney, and Stable Diffusion with platform-specific optimization, style keywords, negative prompting, aspect ratio guidance, and a structured prompt engineering framework.

A Claude Code plugin marketplace from Trail of Bits providing skills for AI-assisted security analysis, testing, and development workflows.

Design effective crewAI agents via the Role-Goal-Backstory framework — specific roles like "Senior Data Researcher specializing in {topic}", outcome-focused goals with quality standards, expertise/values in the backstory (never implementation details). Covers execution knobs (`max_iter`, `max_rpm`, `max_execution_time`), tool assignment (fewer-focused-tools wins), `function_calling_llm` for cost savings, delegation, planning, `allow_code_execution`, guardrail callbacks, knowledge sources, and YAML-first configuration.