Browse the library

MCP server that indexes your repo across many languages and frameworks so agents get answers from the index, cutting input tokens when reviewing pull requests.

A visual workspace where your agent builds its own UI, with widgets and views wired to APIs, local files and MCP servers, taught through a skill.

A toolkit of three Agent Skills to mine workflows, personalize existing skills, and generalize personal skills for public sharing.

Provides a methodology and skills to build products with AI agents that remember, by turning documentation into a navigable knowledge graph.

Provides reusable agentic loops as open-standard Agent Skills for tasks like autoresearch, data analysis, and code optimization.

Turns documents dropped in a raw/ folder into a cross-linked markdown wiki using a multi-agent newsroom where the writer and reviewer are separate agents.

An MCP server and Claude Code plugin that asks Codex, Gemini, Grok and OpenRouter models for second opinions or consensus reviews of plans and code.

A skill that lets Claude Code or Codex create 3Blue1Brown-style Manim animations with local voiceover, subtitles and a browser viewer for feedback.

Enables your AI agent to search, recommend, book, and monitor prices for hotels globally, with Alipay payment and order tracking.

A plugin of engineering-discipline skills for Claude Code, Codex and Gemini CLI covering research, quality gates, testing, deployment and evidence before claiming done.

Audits autoresearch papers for self-consistency and fabrication using 61 integrity signals, providing a deterministic verdict.

A curated Chinese-language awesome list of open-source legal AI tools and skills for lawyers, organised by task with usage and data-path notes.

Lets your AI agent edit raw video clips into a finished, beat-synced montage using natural language prompts without editing software.

Generates viral Xiaohongshu posts including vertical cards, captions, and hashtags, from topics or GitHub project links.

A skill/plugin that lets your coding agent make videos locally, with motion graphics, voice-over, music, captions and exports, without API keys or uploads.

Offers a collection of 23 agent skills and safety hooks for AI-driven software development across multiple coding agents.

RecallLoom provides persistent project memory for AI agents, saving context, decisions, and next steps in local project files without a database.

This subagent orchestrates tasks between Claude, Gemini, and GPT based on task scope, routing large-scale analyses to Gemini and focused work to GPT, with Claude review.

Unslop is a plugin that strips AI-isms from generated text, making it sound more human while preserving code, URLs, and formatting.

Flompt is a visual AI prompt builder that lets users decompose prompts into a flowchart, edit them, and recompile into Claude-optimized XML.

Yume provides a native desktop UI for Claude Code, offering advanced orchestration, live thinking, background agents, and multi-provider support.

Coder Eval is an agent-agnostic framework for evaluating and benchmarking AI coding agents and their skills using declarative YAML tasks and sandboxed execution.

This is a curated directory of over 100 sub-agent prompts and MCP servers for Claude Code, searchable by category and keyword.

codeArbiter provides an open-source governance layer and hard gates for AI coding agents to enforce tests, reviews, security checks, and durable project context.