A curated list of awesome AI agents, frameworks, tools, and resources.
English | 中文
- A List of Claude Code Agents by Really Him - Community-submitted collection of Claude Code sub-agents and agent frameworks.
- Agentics by GitHub Next - Sample pack of reusable GitHub Agentic Workflows for repository maintenance, code review, CI analysis, security scanning, documentation updates, and more.
- Get Shit Done by GSD - A lightweight meta-prompting, context engineering, and spec-driven development system that solves context rot with fresh subagent sessions, parallel wave execution, and atomic git commits. Supports 15+ AI coding runtimes.
- gstack by Garry Tan - Garry Tan's virtual engineering team for Claude Code — 23 opinionated slash-command skills covering CEO review, design, engineering, QA, security audit, and release, with real browser automation and multi-agent coordination.
- Superpowers by Jesse Vincent - An agentic skills framework and software development methodology for coding agents, featuring TDD, subagent-driven development, brainstorming, and systematic debugging skills. Supports Claude Code, Codex, Cursor, Gemini CLI, and more.
- Claw-Eval by Bowen Ye & Lei Li (PKU/HKU) - Trustworthy evaluation benchmark for autonomous agents, covering general, multi-turn, and multimodal tasks with safety-aware scoring.
- ProjDevBench by Pengrui Lu - Benchmark for evaluating AI coding agents on end-to-end project development, with dual evaluation (Online Judge + LLM code review), Docker isolation, and support for Claude Code, Cursor, Gemini CLI, Codex, Augment, and Copilot.
- Vellum LLM Leaderboard by Vellum - Latest public benchmark performance for SOTA models, featuring results from GPQA Diamond, AIME 2025, SWE-Bench, Humanity's Last Exam, ARC-AGI 2, MMMLU, and more.
- AI HOT by Virxact - Curated AI news aggregator with daily digests and trend tracking, offering REST API, RSS, and SKILL.md integration for coding agents.
- Andrej Karpathy Skills by Jiayuan Zhang - Karpathy-inspired CLAUDE.md guidelines addressing LLM coding pitfalls with four principles: think before coding, simplicity first, surgical changes, and goal-driven execution.
- Anthropic Skills by Anthropic - Anthropic's official skill collection for Claude Code, including skill-creator (create, evaluate, and optimize skills with benchmarking), and more.
- Awesome Agent Skills by VoltAgent - Curated collection of 1100+ agent skills from official dev teams and the community, compatible with Claude Code, Codex, Gemini CLI, Cursor, and more.
- Matt Pocock Skills by Matt Pocock - Small, composable agent skills for real engineering — covers grilling sessions, TDD, diagnosis loops, codebase architecture improvement, triage, and domain-driven documentation. Compatible with Claude Code, Codex, and other coding agents.
- Obsidian Skills by Steph Ango - Agent skills for Obsidian, teaching your coding agent to use Obsidian Markdown, Bases, JSON Canvas, CLI, and Defuddle. Follows the Agent Skills specification, compatible with Claude Code, Codex CLI, OpenCode, and more.
- Startup Claude Skills by Eric Tech - Reusable Claude Code skills for startups, covering bug fixing, feature development, PR review, automated review-fix loops, and headless browser QA testing.
- Agentic Design Patterns by xindoo - Systematic introduction to AI agent design patterns, covering prompt chaining, routing, parallelization, reflection, tool use, planning, multi-agent collaboration, memory management, learning and adaptation, MCP, and more.
- Building Effective Agents by Anthropic - Anthropic's guide on building LLM agents with simple, composable patterns — covers augmented LLM, prompt chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer, and autonomous agents, with practical advice on tool engineering.
- Learn Claude Code by CrazyBoyM / shareAI-lab - 19 chapters across 4 stages, from the minimal agent loop to a multi-agent platform, covering tool use, subagents, skills, context compaction, permissions, hooks, memory, error recovery, task scheduling, team protocols, worktree isolation, and MCP.
- YouTube Tutorials by Owain Lewis - Self-contained companion repos for YouTube tutorials on practical AI automation patterns, covering RAG, agent teams, background agents, multi-agent systems, spec-driven development, and more.
Contributions welcome! Read the contribution guidelines first.
