Cactus Compute Resources

A builder-focused guide to academic-research-skills, a Claude Code skill suite for literature review, writing, peer review, revision, and finalization.

A practical guide to addyosmani/agent-skills, a Markdown skill pack that gives AI coding agents specs, plans, tests, reviews, and release gates.

The official OpenAI Cookbook is a comprehensive collection of examples, guides, and best practices for using the OpenAI API. Covers GPT-4, ChatGPT, embeddings, fine-tuning, and more.

A builder-focused guide to the wshobson/agents repository, a large multi-harness library of agents, skills, plugins, and commands.

A practical overview of the Agent2Agent protocol, why it matters for multi-agent systems, and how builders should evaluate it.

A practical guide to Awesome LLM Apps, the GitHub repository of runnable AI agent, RAG, MCP, voice agent, and fine-tuning templates.

A practical guide to the Machine-Learning-Interviews repository for ML, AI engineering, deep learning, and system-design interview prep.

A Chinese-language open-source tutorial series for hands-on LLM programming and practice.

A Chinese-language guide for fine-tuning and deploying open LLMs and multimodal models on Linux.

The repository teaches a Transformer-based language model workflow in PyTorch, covering data download, preprocessing, model building, training, checkpoints, text generation, and post-training methods such as SFT, reward modeling, PPO, DPO, and GRPO/RLVR. The GitHub summary reports an MIT license, roughly 5.6k stars, and a companion documentation site.

The repository is a curated collection of practical AI engineering projects covering OCR systems, RAG apps, AI agents, audio applications, multimodal workflows, and production-oriented examples. The GitHub summary reports about 2.1k stars, MIT licensing, and recent activity in June 2026.

A public GitHub reference collection of extracted system prompts from major AI tools, useful for studying prompt design, agent behavior, and safety patterns.

A builder-focused guide to the agentic-awesome-skills catalog, including what it is, when to use it, evaluation checks, and adoption risks.

A beginner-friendly large language model application development course from Datawhale China, reviewed as an OpenTools learning resource.

A practical guide to Google Labs Code DESIGN.md, the Markdown-and-YAML format that gives coding agents structured design-system context.

Get Codex running in under 5 minutes. Covers CLI installation on every platform, authentication with ChatGPT or API keys, your first tasks, the desktop app, Codex Cloud, and IDE extensions.

A practical GitHub resource for developers learning LLM, RAG, and AI agent engineering through hands-on tutorials and examples.

A community-created collection of custom agents, instructions, skills, hooks, workflows, and plugins to supercharge your GitHub Copilot experience. 31K+ GitHub stars.

A practical guide to using Graphify to turn code, schemas, scripts, documents, and media into a queryable graph for AI coding workflows.

fastbook is the public notebook companion for the fastai deep learning book and MOOC, covering practical AI with fastai and PyTorch.

A practical guide to Hallmark, an open-source design skill that helps coding agents produce less generic UI work.

Master Codex CLI: interactive vs exec mode, key flags (--full-auto, --sandbox, --model), config.toml setup, slash commands, and remote WebSocket sessions.

A visual, example-driven guide to Claude Code covering basic concepts through advanced agents, with copy-paste templates and quizzes. Synced with every Claude Code release.

A comprehensive guide to the Codex Skills and Plugins system - how skills are defined, stored, invoked, bundled into plugins, and configured for OpenAI agents.

Set up Claude Code in VS Code, JetBrains IDEs, the Desktop app, or Web mode. Compare surfaces and find the right one for your workflow.

Anthropic's official collection of Jupyter notebooks and code recipes demonstrating practical Claude usage patterns including agents, tool use, RAG, and multimodal capabilities.

A practical guide to ARIS, the Auto-Research-In-Sleep workflow kit for autonomous ML research agents and Markdown-only skills.

A high-signal GitHub reference collection for studying system prompts, internal tool specs, and model patterns across popular AI products.

AI-For-Beginners is Microsoft’s twelve-week, twenty-four-lesson curriculum for learning AI fundamentals with structured lessons and hands-on practice.

Anthropic Skills is a public reference repository for Claude and Agent Skills, with examples, templates, and patterns for reusable agent workflows.

A practical course and code repository for learning how to design, develop, deploy and iterate on production-grade machine learning applications.

A comprehensive collection of Jupyter notebooks and code recipes showcasing effective ways to use Claude for classification, RAG, summarization, tool use, vision, and more.

Community-contributed collection of GitHub Copilot instructions, agents, skills, and configurations to help you get the most out of GitHub Copilot in your workflow.

A public companion repository with supporting materials for Chip Huyen’s AI Engineering book and practical AI engineering study paths.

A curated collection of 100+ specialized Claude Code subagent configurations covering code review, testing, DevOps, database, security, and more.

A practical guide to using Datawhale China’s hello-agents course as a structured learning path for AI agent principles, examples, and implementation patterns.

A guide to Ed Donner’s agents repository, which supports a six-week Agentic AI Engineering course using OpenAI Agents SDK, CrewAI, LangGraph, AutoGen, and MCP.

A curated OpenAI repository showing Codex plugin examples, manifests, skills, app integrations, commands, hooks and MCP-backed workflow patterns.

Claude Code Game Studios packages agents, workflow skills, hooks, rules, and templates for structured AI-assisted game development in Claude Code.

🎭 211 个即插即用的 AI 专家角色 — 支持 Hermes Agent/Claude Code/Cursor/Copilot 等 16 种工具，覆盖工程/设计/营销/金融等 18 个部门。含 46 个中国市场原创智能体（小红书/抖音/微信/飞书/钉钉等）

A builder-focused guide to Andrej Karpathy's Neural Networks: Zero to Hero course repository for learning neural networks from first principles.

The most comprehensive toolkit for Claude Code -- 135 agents, 35 curated skills, 42 commands, 176+ plugins, 20 hooks, 15 rules, 7 templates, 14 MCP configs, 26 companion apps, 52 ecosystem entries, and more.

GEO-first SEO skill for Claude Code. Comprehensive AI search optimization for any website — citability scoring, AI crawler analysis, brand authority, schema markup, platform-specific optimization, and PDF reports. If you want learn how to s

Claude Code Tips is a practical collection of 45 workflow tips, scripts, commands, and patterns for using Claude Code more effectively.

learn-claude-code is a 0-to-1 tutorial for building a Claude Code-like agent harness with tools, context, permissions, and progressive sessions.

Anthropic’s interactive prompt engineering tutorial teaches Claude prompting through 9 chapters, exercises, examples, and a Google Sheets learning path.

Microsoft’s free Generative AI for Beginners course teaches 21 lessons for building AI apps with Python, TypeScript, Azure OpenAI, GitHub Models, and OpenAI.

A curated GitHub index of official, free, written AI and machine-learning course notes from universities, filtered for prose quality and primary-source links.

A practical guide to the LLMs-from-scratch repository, the official PyTorch code companion for learning GPT-style models step by step.

A practical guide to mattpocock/skills, a high-signal collection of composable skills for Claude Code, Codex, and coding agents.

A practical guide to Microsoft ML-For-Beginners, a free 12-week curriculum for classic machine learning with Python and scikit-learn.

Scientific Agent Skills is a large open-source library of ready-to-use skills for research, science, engineering, finance, and writing agents.

A source-backed guide to the 500 AI Agents Projects repository, a curated collection of practical AI-agent use cases and open-source implementation links.

A curated reference to Arindam Majumder’s awesome-ai-apps repository, a large collection of practical RAG, agents, voice, MCP, memory, and workflow examples.

Agent Skills is an Apache-2.0 specification and documentation project for packaging reusable AI-agent capabilities as portable skill folders.

Connect Claude Code to databases, browsers, APIs, and external tools using the Model Context Protocol. Setup guides for popular servers and troubleshooting.

Side-by-side comparison of Claude Code, Cursor, and GitHub Copilot. Features, codebase depth, pricing, and when to choose each tool.

A practical guide to Anthropic's official Claude Code plugin directory, its repository structure, install command, trust model, and use cases.

A curated Claude Skills resource for builders who want reusable instruction packages, plugin examples, app-connection workflows, and practical Claude Code customization patterns from the ComposioHQ repository.

marketingskills is a GitHub resource library of marketing-focused Agent Skills for Claude Code, Codex, Cursor, Windsurf, and AI agents.

A Packt companion repository and learning resource for building practical AI agents with notebooks, provider setup, and architecture examples.

A source-backed guide to Cloudflare Skills, Cloudflare’s public skills repository for teaching agents how to build on Cloudflare’s developer platform.

A source-backed guide to human-writing, a Chinese writing and editing skill that helps AI-generated Chinese text read like a specific person rather than generic model prose.

A source-backed guide to claude-seo, a Claude Code skill collection for technical SEO, E-E-A-T, schema, GEO/AEO, backlinks, local SEO, Google APIs, and reporting.

A source-backed guide to Google’s public Agent Skills repository for Google Cloud, Gemini, Agent Platform, GKE, BigQuery, RAG, monitoring, logging, and agent operations.

A practical reference to BuilderIO Skills, the open-source collection of composable skills for coding agents such as Codex, Cursor, OpenCode, GitHub Copilot, and Claude Code.

A source-backed guide to Andrej Karpathy’s micrograd repository, a tiny educational autograd engine and neural-network library with a PyTorch-like API.

Google Gemma Cookbook is a collection of examples, guides, and notebooks for working with Google’s Gemma open model family.

Open-source course for building a second brain AI assistant with LLMs, RAG, agents, fine-tuning, LLMOps, and production AI systems practices.

A practical resource for evaluating the wondelai-skills collection of reusable agent skills for Claude Code and agentskills.io-compatible coding agents.

nanoGPT is Andrej Karpathy’s compact GPT training and finetuning resource for builders who want to understand medium-sized GPTs through runnable PyTorch code, Shakespeare quickstarts, and GPT-2 reproduction notes.

Practical patterns, starters & CLI tools for loop engineering with AI coding agents. Design systems that prompt and orchestrate agents (inspired by Addy Osmani and Boris Cherny). Includes loop-audit, loop-init, loop-cost. OpenTools classifies it as a developer resource because it teaches loop engineering patterns for AI coding agents.

Use TorchCode to practice PyTorch operators, attention, GPT-style components and ML interview problems in a Jupyter-based training environment today.

Study Claude Code system prompts, builtin tool descriptions, sub-agent prompts and utility prompts with this maintained GitHub reference collection today.

impeccable is a source-backed AI engineering resource. The design language that makes your AI harness better at design.

A collection of 100+ specialized Claude Code subagents covering a wide range of development use cases OpenTools classifies it as a developer resource because it is a Claude extension and agent workflow hub, not a single tool or model.

非官方 Kimi Slides Skill：让 AI Agent 生成可编辑 PPTD + PPTX，并附带本地浏览器编辑器 Unofficial Kimi Slides skill for AI agents — generate editable PPTD + PPTX with a local browser editor OpenTools classifies it as a developer resource because it is an unofficial Kimi Slides skill package for AI agents.

Repeatable agents-plus-code workflows, packaged as one skill, stamped into any repo. Deterministic Python owns the graph; coding agents are bounded nodes inside it. OpenTools classifies it as a practical AI-builder resource with source-backed setup and learning notes.

LLM Zoomcamp - a free online course about real-life applications of LLMs. In 10 weeks you will learn how to build an AI system that answers questions about your knowledge base. OpenTools classifies it as a practical AI-builder resource with source-backed setup and learning notes.

RealReplicaBench: Benchmarking Long-Horizon Agents in High-Fidelity, Stateful, and Reproducible Replicas of Real Online Services OpenTools classifies it as an AI-agent evaluation resource for builders testing realistic long-horizon workflows.

A practical guide to the read-only AI skill for finding domain names, checking availability in bulk, and comparing renewal prices from registrars.

RAG_Techniques is a large tutorial repository for advanced Retrieval-Augmented Generation patterns, notebooks, scripts, and evaluation examples.

reverse-skill is a security research skill router pack for Claude Code, Kiro, Cursor, Cline, and similar AI coding clients. It focuses on authorized reverse engineering and penetration-testing workflows.

Mythos Claude Skill Forge is a community Claude skills and plugin pack for experimenting with lenses, subagents, hooks, and agent behavior design.

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术 OpenTools classifies it as an AI systems learning resource for builders working below the API layer.

Anthropic Cybersecurity Skills is a large community skill library that maps AI-agent cybersecurity workflows to MITRE, NIST, ATLAS, D3FEND, and AI RMF frameworks.

The Gemini API Cookbook is Google Gemini’s hands-on example repository for quickstarts, notebooks, agents, multimodal use cases, and practical API patterns.

compound-engineering-plugin is a source-backed AI engineering resource. Official Compound Engineering plugin for Claude Code, Codex, Cursor, and more

A single hub to find Claude Skills, Agents, Commands, Hooks, Plugins, and Marketplace collections to extend Claude Code, Claude Desktop, Agent SDK and OpenClaw OpenTools classifies it as a developer resource because it is a Claude extension and agent workflow hub, not a single tool or model.

A Gemini 2.5 and LangGraph quickstart for building full-stack agent apps. OpenTools classifies it as a developer resource because it is source-backed example code.

Product Management skills framework built on battle-tested methods for Claude Code, Cowork, Codex, and AI agents. OpenTools classifies it as a resource because it is a reusable skills framework rather than a standalone software product.

Automate code review, issue triage, and PR creation with Claude Code in GitHub Actions, GitLab CI, and headless mode.

Master context management, plan mode, auto mode, effort levels, git workflows, voice input, custom status lines, and autonomous work patterns.

NVIDIA Nemotron Developer Asset Hub is a source-backed GitHub resource for AI builders. OpenTools verified the upstream repository, public metadata, and latest activity before publishing this guide.

The ultimate configuration framework for Claude Code with 28 subagents, 119 skills, 60 commands, and AgentShield security scanner. 168K GitHub stars.

A practical guide to Awesome Claude Code, a curated GitHub list of Claude Code skills, hooks, slash commands, agent orchestrators, apps, and plugins.

A practical guide to the qiaomu Anything to NotebookLM Claude Skill for converting web, video, PDF, Markdown, and search sources into NotebookLM-ready outputs.

A practical guide to the pm-skills repository, a marketplace-style collection of agentic product-management skills, commands, and plugins for discovery, strategy, launch, and growth workflows.

A practical guide to using Matt Pocock's public Claude Code skills repository as a reference library for reusable local AI coding workflows.

A practical guide to the claude-code-best-practice repository, covering subagents, commands, skills, hooks, memory, MCP, and workflow orchestration for Claude Code users.

Happy LLM From-Scratch Course is a source-backed GitHub resource for AI builders. OpenTools verified the upstream repository, public metadata, and latest activity before publishing this guide.

Web Quality Skills for AI Agents is a source-backed GitHub resource for builders using agent skills to audit Lighthouse, Core Web Vitals, accessibility, SEO, and frontend quality checks.

A GitHub collection of specialized AI agency agent profiles, workflows, and deliverable patterns for builders designing multi-agent service teams.

A practical guide to diagram-design, the Claude Code plugin and agent-skill resource for creating editorial HTML and SVG diagrams without Mermaid defaults.

The official Model Context Protocol servers repository, useful as a reference catalog for MCP server implementations and integration patterns.

A Claude Code skill-pack resource for Chinese web-novel workflows, covering research, outline building, drafting, style cleanup, and cover flow.

Caveman is a Claude Code skill that intentionally compresses communication style to reduce token use while keeping instructions understandable for coding work.

superpowers-zh is a Chinese enhanced edition of the Superpowers AI coding workflow, with localized material and original skills for AI coding tools.

Google Cloud’s generative-ai repository collects sample code, notebooks, and examples for Gemini, Vertex AI, and related cloud AI workflows.