Browse agent skills

AI-powered coding toolbar for any web application. Select elements, capture context, and let AI modify your codebase in real-time.

qlabs-style Voice AI Platform API

Fusioni Chat SDK for React applications

NPM installer for the requirement-assistant Codex skill

Media pipelines - Declarative YAML-based workflows

XHS data scraping suite for Claude Code & Openclaw. Two skills: TikHub API (multi-platform) + Rnote API (XHS focused). Supports XHS, Douyin, TikTok, Bilibili, Weibo, YouTube, WeChat, Twitter/X.

No description available

Standalone belief and memory engine for Node.js — SQLite, zero dependencies, with a built-in LLM tool facade

铜炉的 skill demo - 演示 skill 的发布和安装流程

Agent Skill for storing and retrieving author information

Convex component for AI and tool cost tracking with flexible attribution

Intelligent orchestration layer for agentic coding — 50 skills, smart skill loading, persistent memory, 70% cost reduction, enforced workflows, and 10+ platform support.

Sends Polymarket alerts to a Telegram group via bot

AI Configuration Manager CLI - Manage AI tool configurations globally and per-project

A thin Playwright CLI wrapper with stealth, extension loading, and persistent profile support.

“Generative AI era” survival. A simple encoder with Table Rotation.

lower-engine-parser is a frontend skill indexed by Skillful.sh. Package: lower-engine-parser.

Opinionated scaffolding CLI to enable AI-agents to manage TypeScript monorepos deterministically, with guided architecture. Leading by example.

Baidu Netdisk CLI tool written in TypeScript

n8n community node for YCloud API integration

Meshtally Node.js SDK — Smart Payment Routing & Orchestration

Self-evolving deep research system for Claude Code. Zero API keys.

B2B software vendor evaluation skill for Claude Code — domain-expert questions, vendor AI agent conversations, evidence-based scoring

Official Node.js/TypeScript SDK for NexaAPI — AI image generation, video generation, text-to-speech, and more. Supports Flux, Stable Diffusion, Kling, Veo, ElevenLabs, and 56+ AI models.

Maestro agent loop, transport, types, and sandbox primitives — the engine behind all Maestro interfaces

TruContext memory for OpenClaw — install once, works forever

`hobonos` is a conversational router. Routes own pages, pages own components, and each incoming message resolves to the component the user is trying to interact with.

Abstract layer for LLM completion supporting multiple providers

A lightweight HTTP client class for Node.js and Browser, to make API calls tasty and easy.

Unified LLM API with automatic model discovery and provider configuration