Browse agent skills

Extract text from images using macOS Vision Framework OCR

Context engine for AI coding agents — semantic code search via CLI

Voice message transcription for OpenClaw via Groq or OpenAI Whisper API

OpenCode Bridge - Use OpenCode AI models in Claude Code

Vercel AI SDK provider for Cursor (using cursor-agent)

Fast, cross-platform multi-account manager for Claude Code

Trust enforcement middleware for AI agents on Celo

A cron parsing and iteration library inspired by cron-parser, powered by date-fns.

Structured LLM metadata standard for Node.js packages — llm.package.json and llm.package.txt

Multi-provider AI chat interface supporting 9+ LLM providers (Anthropic, OpenAI, xAI, Cohere, Mistral, Perplexity, Gemini, HuggingFace, Groq)

High-performance streaming Markdown text splitter for LLM pipelines and RAG systems. Zero sequence loss for code blocks, tables, links, and videos. 5 built-in strategies + custom. Zero dependencies.

Enterprise-grade monitoring SDK - Smarter than DataDog with adaptive sampling, circuit breaker, system metrics, error tracking, and zero dependencies

Tavily web tools plugin for OpenClaw — exposes tavily_search, tavily_extract, tavily_crawl, tavily_map, and tavily_research as agent tools

Extract design system patterns from HTML files into structured Markdown references — colors, typography, spacing, animations, components

n8n node for BaseVN Workflow API

A complete book-writing pipeline powered by multi-agent AI — canonical short context, adversarial review, and progressive summaries keep every chapter consistent.

AgentStudio - AI-powered agent workspace with Claude integration. One command to start your AI agent studio.

n8n node for BaseVN Office API

EU Infinite Scrolling Compliance SDK - Drop-in React pagination for GDPR compliance

TypeScript client for the Gotenberg document conversion API.

n8n node for BaseVN HRM API

Production-ready guides for building AI agents with Claude Agent SDK in Python and TypeScript - complete coverage of official docs plus advanced patterns (subagents, hooks, batch processing)

Multi-agent AI orchestration framework — coordinate Claude, Gemini, Codex, Amp, and OpenCode as a unified dev team

Skills SDK for JavaScript

Security middleware for RAG pipelines — detect adversarial hallucination attacks before they reach your LLM.

TypeScript implementation of BAML's Schema-Aligned Parsing (SAP) algorithm using TypeBox

Agent framework SDK powered by minns-sdk memory layer

Core types and utilities for building Poncho skills

Lightweight Playwright fixture that automatically tracks and closes extra pages and contexts created during tests — even when they fail, timeout, or retry

RAG chunk reordering, retrieval reranking, and context packing to mitigate lost-in-the-middle in LLM context windows. Drop-in reordering for LangChain, LlamaIndex, Vercel AI, and any RAG pipeline.