Market Radar
What each category rewards right now — GitHub candidates ranked by stars × recency. Regenerated by the discovery engine.
Claude Code Skills & Subagents
84 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| affaan-m/ECC | 225,688 | 225,688 | The agent harness performance optimization system. Skills, instincts, memory, security, an |
| multica-ai/andrej-karpathy-skills | 187,207 | 140,788 | A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's ob |
| farion1231/cc-switch | 112,838 | 112,838 | A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, |
| JuliusBrussee/caveman | 82,928 | 82,928 | 🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens b |
| safishamsi/graphify | 77,119 | 77,119 | AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Tu |
| ruvnet/ruflo | 62,855 | 62,855 | 🌊 The leading agent meta-harness. Deploy intelligent multi-player swarms, coordinate auton |
| santifer/career-ops | 58,398 | 58,398 | AI-powered job search system built on Claude Code. 14 skill modes, Go dashboard, PDF gener |
| hesreallyhim/awesome-claude-code | 47,910 | 47,910 | A hand-picked collection of the finest of resources for the most awesome of agents, Claude |
| sickn33/antigravity-awesome-skills | 42,296 | 42,296 | Installable GitHub library of 1,800+ agentic skills for Claude Code, Cursor, Codex CLI, Ge |
| danny-avila/LibreChat | 40,238 | 40,238 | Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Re |
Agent Loops & Orchestration
86 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| earendil-works/pi | 67,540 | 67,540 | AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI |
| code-yeongyu/oh-my-openagent | 64,779 | 64,779 | omo/lazycodex: The coding agent for tokenmaxxers;the one and only agent harness for comple |
| Yeachan-Heo/oh-my-claudecode | 37,385 | 37,385 | Teams-first Multi-agent orchestration for Claude Code |
| wshobson/agents | 37,496 | 36,781 | Multi-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, Git |
| jnMetaCode/agency-agents-zh | 16,575 | 16,385 | 🎭 266 个即插即用的 AI 专家角色 — 支持 Hermes Agent/Claude Code/Cursor/Copilot 等 18 种工具,覆盖工程/设计/营销/金融等 |
Cursor Rules & Workflows
85 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| ComposioHQ/awesome-claude-skills | 66,746 | 56,779 | A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI wo |
| sickn33/antigravity-awesome-skills | 42,296 | 42,296 | Installable GitHub library of 1,800+ agentic skills for Claude Code, Cursor, Codex CLI, Ge |
| PatrickJS/awesome-cursorrules | 40,208 | 35,274 | 📄 Configuration files that enhance Cursor AI editor experience with custom rules and beha |
| czlonkowski/n8n-mcp | 22,128 | 22,128 | A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you |
| jnMetaCode/agency-agents-zh | 16,526 | 16,399 | 🎭 266 个即插即用的 AI 专家角色 — 支持 Hermes Agent/Claude Code/Cursor/Copilot 等 18 种工具,覆盖工程/设计/营销/金融等 |
| refly-ai/refly | 7,417 | 5,046 | The first open-source agent skills builder. Define skills by vibe workflow, run on Claude |
| sanjeed5/awesome-cursor-rules-mdc | 3,551 | 2,986 | Curated list of awesome Cursor Rules .mdc files |
| vanzan01/cursor-memory-bank | 3,049 | 2,634 | A modular, documentation-driven framework using Cursor custom modes (VAN, PLAN, CREATIVE, |
| rohitg00/pro-workflow | 2,586 | 2,546 | Claude Code learns from your corrections: self-correcting memory that compounds over 50+ s |
| stravu/crystal | 3,095 | 1,898 | (Crystal is now Nimbalyst) Run multiple Codex and Claude Code AI sessions in parallel git |
Spec & PRD Templates for AI Coding
70 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| github/spec-kit | 117,774 | 117,321 | 💫 Toolkit to help you get started with Spec-Driven Development |
| Fission-AI/OpenSpec | 58,570 | 58,570 | Spec-driven development (SDD) for AI coding assistants. |
| Pimzino/spec-workflow-mcp | 4,255 | 4,255 | A Model Context Protocol (MCP) server that provides structured spec-driven development wor |
| centminmod/my-claude-code-setup | 2,456 | 2,437 | Shared starter template configuration and CLAUDE.md memory bank system for Claude Code |
| KhazP/vibe-coding-prompt-template | 2,582 | 1,934 | Templates and workflow for generating PRDs, Tech Designs, and MVP and more using LLMs for |
| zhu1090093659/spec_driven_develop | 912 | 905 | Spec-driven development workflow for AI coding agents: architecture-first planning, task d |
| Gentleman-Programming/agent-teams-lite | 1,245 | 850 | Spec-Driven Development with AI Sub-Agents. An orchestrator + 9 specialized sub-agents for |
| shotgun-sh/shotgun | 684 | 607 | Spec Driven Development 🤠 Write codebase-aware specs for AI coding agents so they don't de |
| ThibautBaissac/rails_ai_agents | 619 | 547 | Specialized AI skills, agents, rules and hooks for modern Rails AI driven-development + Sp |
| josix/awesome-claude-md | 481 | 432 | Curated collection of exemplary claude.md files and onboarding patterns from public GitHub |
MCP Server Starters
88 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| modelcontextprotocol/servers | 88,025 | 88,025 | Model Context Protocol Servers |
| punkpeye/awesome-mcp-servers | 90,252 | 87,852 | A collection of MCP servers. |
| modelcontextprotocol/python-sdk | 23,521 | 23,431 | The official Python SDK for Model Context Protocol servers and clients |
| modelcontextprotocol/typescript-sdk | 12,783 | 12,783 | The official TypeScript SDK for Model Context Protocol servers and clients |
| yzfly/Awesome-MCP-ZH | 7,387 | 7,387 | MCP 资源精选, MCP指南,Claude MCP,MCP Servers, MCP Clients |
| modelcontextprotocol/registry | 6,979 | 6,926 | A community driven registry service for Model Context Protocol (MCP) servers. |
| hangwin/mcp-chrome | 12,037 | 6,065 | Chrome MCP Server is a Chrome extension-based Model Context Protocol (MCP) server that exp |
| getsentry/XcodeBuildMCP | 6,027 | 5,935 | A Model Context Protocol (MCP) server and CLI that provides tools for agent use when worki |
| mobile-next/mobile-mcp | 5,362 | 5,341 | Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators, |
| FlorianBruniaux/claude-code-ultimate-guide | 5,317 | 5,317 | The most comprehensive Claude Code guide: agentic workflows, hooks, skills, MCP servers, q |
Eval Suites & LLM Testing
90 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| openai/evals | 18,829 | 13,837 | Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of b |
| modelscope/evalscope | 3,027 | 3,027 | A streamlined and customizable framework for efficient large model (LLM, VLM, AIGC) evalua |
| itayinbarr/little-coder | 1,698 | 1,666 | A harness optimized to smaller LLMs |
| suyoumo/ClawProBench | 809 | 794 | ClawProBench is a live-first benchmark harness for evaluating LLM agents in the OpenClaw |
| verifywise-ai/verifywise | 315 | 315 | Complete AI governance and LLM Evals platform with support for EU AI Act, ISO 42001, NIST |
| Shiyao-Huang/awesome-agent-evolution | 170 | 170 | Open survey and evidence map for AI agent evolution, self-evolving agents, memory, skills, |
| strands-agents/evals | 153 | 151 | A comprehensive evaluation framework for AI agents and LLM applications. |
| alibaba-damo-academy/MedEvalKit | 245 | 149 | MedEvalKit: A Unified Medical Evaluation Framework |
| openclaw/shellbench | 124 | 120 | The agent benchmark that scores the full stack — harness, config, and model — not just the |
| agentscope-ai/PawBench | 83 | 82 | A benchmark for evaluating LLM × harness performance. |
Agent Teams & Role Bundles
90 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| msitarzewski/agency-agents | 126,928 | 125,954 | A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas |
| tanweai/pua | 18,623 | 17,443 | 你是一个曾经被寄予厚望的 P8 级工程师。Anthropic 当初给你定级的时候,对你的期望是很高的。 一个agent使用的高能动性的skill。 Your AI has be |
| jnMetaCode/agency-agents-zh | 16,593 | 16,402 | 🎭 266 个即插即用的 AI 专家角色 — 支持 Hermes Agent/Claude Code/Cursor/Copilot 等 18 种工具,覆盖工程/设计/营销/金融等 |
| jnMetaCode/agency-orchestrator | 1,637 | 1,575 | 🚀 One sentence → multi-AI-role collaboration → complete plan in minutes. Built on the agen |
| operand/agency | 487 | 446 | A fast and minimal framework for building agentic systems |
| Anas-Khan93/ai-agency-agents | 489 | 442 | Open Agency is an open-source AI workforce featuring specialized agents for development, g |
| linkerlin/PUAX | 314 | 314 | 用来驯服AI Agent的效果绝佳的多角色 Prompt! |
| indranilbanerjee/digital-marketing-pro | 206 | 201 | Open-source AI marketing plugin for agencies & in-house teams — 158 skills, 25 specialist |
| dfinke/PSAI | 266 | 179 | High-agency PowerShell AI framework for multi-agent orchestration and autonomous systems e |
| chenhg5/agencycli | 145 | 144 | Lightweight CLI to build self-managing AI agent teams. Define roles, skills & projects in |
Prompt Systems & Goal Loops
89 candidates seen · scanned 2026-07-04
| repo | stars | score | what |
|---|---|---|---|
| gsd-build/get-shit-done | 64,656 | 56,940 | A light-weight and powerful meta-prompting, context engineering and spec-driven developmen |
| gsd-build/gsd-2 | 7,747 | 6,590 | A powerful meta-prompting, context engineering and spec-driven development system that ena |
| cobusgreyling/loop-engineering | 5,330 | 5,330 | Practical patterns, starters & CLI tools for loop engineering with AI coding agents. Desig |
| Meirtz/Awesome-Context-Engineering | 3,221 | 2,804 | 🔥 Comprehensive survey on Context Engineering: from prompt engineering to production-grad |
| EmbeddedLLM/JamAIBase | 1,102 | 1,001 | The collaborative spreadsheet for AI. Chain cells into powerful pipelines, experiment with |
| open-gsd/gsd-pi | 819 | 819 | A powerful meta-prompting, context engineering and spec-driven development system that ena |
| 10xChengTu/harness-engineering | 92 | 82 | Set up and improve harness engineering (AGENTS.md, docs/, lint rules, eval systems, projec |
| vasilyevdm/ai-agent-handbook | 101 | 67 | Comprehensive guide to AI agent engineering: how 30+ frameworks actually work under the ho |
| harsh-aranga/ai-engineering-bootcamp | 68 | 64 | A self-paced bootcamp for engineers building AI into production systems. Starts from funda |
| moely-ai/claude-fable-5-prompt | 57 | 53 | A curated collection of Claude Fable 5 system prompts for developers and researchers. Disc |