Repositories
Recommended GitHub repositories for AI tools, agent frameworks, and open-source implementations.
Last updated: 1 August 2026
agent-skills
20 production-grade engineering workflows packaged as structured skills across the dev lifecycle (Define, Plan, Build, Verify, Review, Ship).
SKILL.md specification
Open standard for portable, reusable AI agent skills.
claude-skills
192+ reusable production skills for Claude Code.
claude-code
Agentic coding tool that lives in your terminal.
claude-code-action
GitHub Action for running Claude Code in CI/CD pipelines.
modelcontextprotocol
Model Context Protocol specification and reference implementations.
claude-cookbooks
Official Anthropic cookbook: runnable notebooks covering tool use, RAG, evaluation, multi-agent patterns, and production guardrails.
awesome-claude-skills
Comprehensive collection of Claude Code skills across domains.
crewAI
Framework for orchestrating role-playing autonomous AI agents with task delegation, memory, and collaborative intelligence.
github/awesome-copilot
Official GitHub repository of community-contributed instructions, agents, skills, and configurations for GitHub Copilot — a practical reference for agentic coding workflows.
f/prompts.chat
Community prompt library (fka Awesome ChatGPT Prompts) covering Claude, Gemini, GPT and more. Self-hostable for org-wide privacy.
Shubhamsaboo/awesome-llm-apps
100+ runnable AI agent and RAG applications covering diverse use cases — clone, customise, and ship.
PatrickJS/awesome-cursorrules
Curated collection of .cursorrules configuration files that tailor Cursor AI editor behaviour across languages and frameworks.
elizaOS/eliza
TypeScript framework for building autonomous agents deployable across Discord, Telegram, Slack and other platforms, with plugin and RAG support.
heygen-com/hyperframes
Write HTML, render video — a TypeScript framework for agent-driven video generation using GSAP, Puppeteer, and ffmpeg.
haystack
Open-source AI orchestration framework for building production-ready LLM pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation.
langgraph
Graph-based orchestration layer from the LangChain team for building stateful, multi-actor agent workflows with fine-grained control over cycles and branching.
mastra
TypeScript-first framework for building AI agents and applications, with built-in evals, MCP support, and workflow primitives — from the team behind Gatsby.
MemoriLabs/Memori
LLM-agnostic agent-native memory infrastructure that turns agent execution and conversation into structured, persistent state for production multi-agent systems.
obra/superpowers
Agentic skills framework and software-development methodology providing reusable shell-based skill primitives for AI coding agents.
punkpeye/awesome-mcp-servers
Curated directory of MCP servers spanning databases, browsers, file systems, and developer tools — the de facto community index for the Model Context Protocol ecosystem.
llama_index
LlamaIndex is a production-grade data framework for building LLM-powered agents and RAG pipelines, with connectors for hundreds of data sources and a modular query engine.
adk-python
Google Agent Development Kit for Python: code-first toolkit for building, evaluating, and deploying multi-agent systems with native A2A protocol support.
timesfm
Pretrained time-series foundation model for zero-shot forecasting.
awesome-claude-code
Curated list of skills, hooks, commands, and agent patterns for Claude Code.
langchain
Framework for building applications with large language models.
dify
Production-ready platform for agentic workflow development: visual pipelines, RAG, model routing, observability.
mem0
Universal memory layer for AI agents: episodic, semantic, and procedural memory stores with retrieval.
autogen
Microsoft's multi-agent conversation framework: define agents with roles, tools, and memory; orchestrate via conversation patterns.
openai-agents-python
OpenAI Agents SDK for building agentic AI applications in Python.
openai-cookbook
Official OpenAI cookbook: example code and guides for GPT models, Assistants, function calling, embeddings, evaluations, and Responses API.
paperclip
Open-source orchestration framework for zero-human companies.
dspy
Framework for programming — not prompting — language models. Declarative modules, compiled prompts, optimisers.