Browse and install thousands of AI Agent skills in the Killer-Skills marketplace. Supports Claude Code, Windsurf, Cursor, and more.
Discover the world's most comprehensive directory for AI Agent Skills and community-driven MCP Servers (Model Context Protocol). Seamlessly find and install high-quality developer integrations for your Claude Desktop, Cursor, or custom AI agents. From cloud operations to API integrations, explore verified open-source MCP practices to supercharge your AI workflows.
component-refactoring is an AI agent skill that refactors high-complexity React components in the Dify frontend. It is activated when `pnpm analyze-component --json` shows a complexity score over 50 or a line count exceeding 300, or upon direct user request for code splitting.
Frontend-testing is a skill that enables AI agents to generate comprehensive tests for frontend components, hooks, and utilities using established conventions and best practices.
orpc-contract-first is a development pattern guiding the implementation of oRPC contract-first APIs in Dify frontend, emphasizing typed contracts and minimal abstractions.
Frontend-code-review is a skill that automates the inspection of frontend code files, applying checklist rules to ensure high-quality code.
supabase-postgres-best-practices is an AI Agent skill containing Supabase's curated guidelines for PostgreSQL database optimization. It provides prioritized rules across 8 categories to automate query review, index implementation, and schema design for improved performance.
hugging-face-datasets is a skill that provides tools for managing datasets on the Hugging Face Hub, including creation, configuration, and SQL-based data manipulation.
hugging-face-evaluation is a skill for managing evaluation results in Hugging Face model cards, supporting multiple methods for adding evaluation data
hugging-face-jobs is a skill for running workloads on fully managed Hugging Face infrastructure, supporting UV scripts, Docker-based jobs, and hardware selection.
hugging-face-model-trainer is a skill for training language models using TRL on Hugging Face infrastructure, covering methods like SFT, DPO, and GRPO
hugging-face-paper-publisher is a comprehensive tool for publishing, managing, and linking research papers on Hugging Face Hub, integrating with arXiv and supporting model/dataset linking.
hugging-face-trackio is an AI Agent skill for tracking and visualizing machine learning training experiments. It provides a Python API for logging metrics, a CLI for retrieving and analyzing logs, and supports real-time dashboard visualization on Hugging Face Spaces.
hugging-face-tool-builder is a skill that creates reusable command line scripts and utilities for using the Hugging Face API, enabling automation and efficient data processing.