Skills

A LaunchDarkly AgentControl skill that creates and manages reusable prompt snippets across configs, enabling consistent prompt components to be shared between different configurations.

A collection of reusable skills for Claude AI and developer tooling, providing modular automation packages for SEO analysis and development workflows.

A UI design skill from the Designer Skills Collection covering typography systems within the ui-design plugin, which provides tools for crafting polished interfaces including layout grids, color systems, typography, responsive design, and data visualization.

Part of the ECC harness-native operator system providing UI demonstration and prototyping patterns for AI agent workflows. Works across Claude Code, Codex, Cursor, Gemini, and other AI coding harnesses with production-ready skills, hooks, and security scanning.

Automates video processing with FFmpeg including cutting, trimming, concatenation, audio mixing, transitions, subtitle handling, and export optimization for YouTube and social media.

A LaunchDarkly AgentControl skill that configures targeting rules for config rollouts, enabling precise control over which users or segments receive specific configurations.

A LaunchDarkly AgentControl skill that manages config variations for A/B testing, enabling experimentation with different prompt configurations to optimize LLM performance.

A LaunchDarkly AgentControl skill that creates and manages projects to organize configs, providing a structured way to group and manage related LLM configurations.

A collection of Claude Code skills covering framework-specific development expertise including FastAPI, Python, and other technologies with best practices.

Enables declarative 3D scene creation in Vue using Three.js, providing reactive components, composables, and helpers for building immersive WebGL experiences.

A collection of auto-generated agent skills derived from the fastest-growing security and AppSec open source projects on GitHub, including skills for cybersecurity practices, penetration testing with AI agents, and security hardening frameworks, installable via the skills CLI.

A financial services plugin skill that enables Claude to clean and prepare raw data in Excel spreadsheets for financial analysis, supporting end-to-end workflows from data gathering through modeling and report generation.

A LaunchDarkly AgentControl skill that updates and deletes configs and manages their lifecycle, providing full CRUD control over LLM prompt configurations.

Guides creation of various content types including blog posts, social media content, marketing copy, and technical documentation using AI-assisted workflows.

A LaunchDarkly skill that creates custom metrics and instruments tracking events for experiments, enabling tailored measurement of feature flag impact beyond built-in metric types.

A skill for building operator-focused monitoring dashboards for platforms like Grafana and SigNoz, structured around real operational questions like health, bottlenecks, and actionable changes rather than vanity metrics.

Provides journal name abbreviation lookup with ISO 4 and MEDLINE support, multi-source cascade (JabRef, AbbrevISO, NLM), BibTeX rewrite with idempotency keys, and atomic cache rebuild for academic citation workflows.

Auto-generated security awareness skill from a bot-curated collection that scans GitHub for trending security and appsec open-source projects, providing installable agent skills for security-related workflows.

A LaunchDarkly AgentControl skill that migrates apps with hardcoded LLM prompts to AgentControl through five stages: extract, wrap, tools, tracking, and evals.

A LaunchDarkly AgentControl skill that attaches LLM-as-a-judge evaluators to configs, enabling automated quality assessment of LLM outputs in production.

A scientific writing skill from a collection of 133 research skills covering cancer genomics, drug discovery, molecular dynamics, and 100+ scientific databases. Transforms AI agents into research assistants for multi-step scientific workflows across biology, chemistry, and medicine.

AI-driven systematic literature review pipeline that performs multi-source retrieval, relevance scoring (1-10), automatic sub-topic grouping, and generates LaTeX+BibTeX output with enforced word count and citation count verification. Supports multilingual output (EN/ZH/JA/DE/FR/ES) with PDF and Word export.

The official WooCommerce monorepo containing the core WordPress e-commerce plugin, extensions, packages, and development tools for building online stores.

A coordination agent skill in the RuFlo platform that orchestrates multi-agent swarms using hierarchical, mesh, or hive-mind topologies with fault-tolerant consensus protocols including Raft, Byzantine, and Gossip for reliable distributed task execution.

A skill for designing and optimizing AI agent action spaces, covering tool definition granularity, observation formatting with structured responses, error recovery contracts, context budgeting strategies, and architecture pattern guidance for ReAct vs function-calling workflows.

A skill that defines the engineering operating model for teams where AI agents generate a large share of implementation output, covering process shifts, agent-friendly architecture requirements, AI-first code review practices, and elevated testing standards.

CloudBase Run backend development skill covering Function and Container modes for long connections (WebSocket/SSE), custom runtime environments, multi-language frameworks, and elastic scaling

Guides you through building MCP (Model Context Protocol) servers with the Node/TypeScript SDK, covering tool and resource registration, Zod validation, stdio vs Streamable HTTP transports, and debugging common SDK API changes.

A task planning agent from Ruflo, a multi-agent AI orchestration platform for Claude Code that breaks big goals into actionable plans and tracks progress, leveraging the goals plugin for structured planning across coordinated agent swarms.

A skill for appending short, stable, project-specific knowledge snippets to AGENTS.md or CLAUDE.md so they are automatically loaded in every AI session, covering build flags, required services, path conventions, and environment variable rules.