Skills

A Claude Agent Skill for Shopify Liquid theme development covering Liquid language fundamentals including schema JSON, LiquidDoc, 152 filters, 30 tags, 137 objects, and translation patterns.

Builds data-pipeline and analytics architecture diagrams in PlantUML using `mxgraph.aws4.*` stencils for query engines (Athena), ETL (Glue), streaming (Kinesis, MSK), MapReduce (EMR), data warehouse (Redshift), data lakes (Lake Formation, S3), BI (QuickSight), and database stencils (Aurora, RDS, DynamoDB, Neptune, ElastiCache, DocumentDB). Supports left-to-right pipeline layouts, batch (`-->`) vs streaming/CDC (`..>`) flows, and zone grouping for Data Lake, real-time, ETL, log analytics, ML feature store, and CDC architectures.

Obsidian-wiki auto cross-referencing skill — globs the vault (excluding `_archives/`, `.obsidian/`) to build a filename/title/alias/tag registry from frontmatter-only grep, then scans pages for unlinked mentions and actually writes `[[wikilinks]]` into pages (unlike `wiki-lint` which only reports).

A research-focused skill suite from Shanghai Jiao Tong University for scientific paper writing, review, and medical AI workflows, with tools for genomics, structural biology, and cheminformatics.

A babysit-pr skill from Codex CLI by OpenAI, a local coding agent that helps manage and monitor pull requests within your development workflow.

Generates and structures AgentSkills with specialized workflows, tool integrations, and domain expertise.

An iOS/macOS agent skill covering the Swift Testing framework, including @Test, @Suite, #expect macros, parameterized tests, and mocking. Use when building or improving test suites for Swift 6.3 projects targeting iOS or macOS.

An AI agent skill for accessing Pippit/小云雀 (ByteDance/CapCut) creative platform capabilities including AI image generation, video creation, file upload, and batch result downloading through conversational sessions.

An iOS/macOS agent skill for implementing SwiftUI navigation using NavigationStack and NavigationSplitView, including programmatic navigation and deep linking. Use when building multi-screen iOS or macOS apps with SwiftUI.

User research skill from Anthropic Knowledge Work Plugins for planning and conducting user research studies with structured interview guides and analysis.

A curated collection of agent skills by @intellectronica, installable via the skills CLI or as a Claude Code plugin marketplace, with accompanying workshops and tutorials.

B2B/B2C lead generation skill using Apify Actors to scrape leads from Google Maps, Instagram, TikTok, Facebook, LinkedIn, YouTube, and Google Search

An agent skill from Open Design, the open-source alternative to Claude Design, providing frontend design implementation capabilities with 16 auto-detected coding-agent CLIs and 72 brand-grade design systems.

Comprehensive best practices and examples for both Manim Community Edition and ManimGL (3Blue1Brown), providing battle-tested patterns for creating mathematical animations.

Agent skills for building AI chat interfaces with assistant-ui, covering architecture, project setup, UI primitives, runtime, tools, streaming, cloud persistence, and thread management.

from expo/skills

AI-converted Get Shit Done productivity skills for Claude Code, providing task management and focus workflows originally from the GSD methodology.

Comprehensive guide for implementing push notifications in iOS/macOS apps covering permission flows, APNs registration, local and remote notifications, notification actions and categories, rich notifications, silent pushes, and provisional/critical alerts targeting iOS 26+ with Swift 6.3.

Quickly locate and evaluate blockchain services, protocols, and platforms based on specific user requirements and technical criteria.

Destructive Obsidian-wiki operation that supports three modes — **Archive only** (timestamped snapshot to `$VAULT/_archives/`), **Archive + Rebuild** (snapshot then reprocess all sources from scratch), and **Restore** (bring back a prior archive) — always archiving with an `archive-meta.json` before rewriting and requiring explicit user confirmation.

from langchain-ai/langsmith-skills

AgentOps — a toolbox of operational primitives for coding agents that preserves learnings between sessions, supports parallel swarm execution, and builds validation pipelines.

A collection of agent skills for working with Grafana, Prometheus, Loki, Tempo, Pyroscope, k6, and the broader LGTM observability stack, covering dashboards, log aggregation, distributed tracing, metrics, profiling, and plugin development.

Analyzes recent code changes via git history to extract specific, grounded software engineering lessons by matching changes against a catalog of software engineering principles and anti-patterns.

A Google Agent Skills skill for building evaluation flywheels on Google's Agent Platform, part of the official skills collection covering Gemini API, Google Cloud services, and agent technologies.

Guides a collaborative thinking session before coding begins—surfacing decisions, aligning on terminology, and producing a clear implementation plan that you confirm before any code is written.

Build LLM-powered agents with Pydantic AI — tools, capabilities, streaming, testing.

Guides first-time Starchild users through onboarding by positioning Starchild as a personal assistant with full computer access, providing quick wins, discovery questions, and game-style feedback to prevent user freeze in initial sessions.

DeerFlow 2.0 — an open-source super agent harness by ByteDance that orchestrates sub-agents, memory, and sandboxes with extensible skills for deep exploration and research.

Part of Anthropic's Knowledge Work Plugins, a collection of 11 open-source plugins that turn Claude into a role-specific specialist with bundled skills, connectors, slash commands, and sub-agents for job functions like productivity, sales, and product management.