genai

langgenius/dify

An all-in-one workspace for building and shipping AI agents — Dify combines agentic workflows, RAG pipelines, and MCP tool support in one collaborative workspace, letting teams design, test and deploy LLM apps with models like GPT, Claude, and DeepSeek. It runs on Dify Cloud or self-hosted, suiting both no-code builders and developers scaling prototypes to production.

mem0ai/mem0

A memory layer that gives AI agents persistent, personalized context — Mem0 lets developers add long-term memory to AI agents and apps, so conversations and context survive across sessions instead of resetting each time. It targets production use, with RAG, state management, and multi-agent memory support, backed by Python and JS SDKs, available as both OSS and hosted platform.

ashishpatel26/500-AI-Agents-Projects

500+ AI agent examples with runnable code across frameworks and industries — A curated catalog of 500+ AI agent projects spanning LangGraph, CrewAI, AutoGen and Agno, organized by industry from healthcare to finance. Each agent under agents/ is self-contained with its own requirements and .env.example, so you can clone and run one in minutes without monorepo setup. Useful for developers picking a framework, researchers surveying the landscape, or students learning agent patterns from working code.

rohitg00/agentmemory

Give Claude, Copilot, Cursor and friends a shared memory — agentmemory adds persistent memory to coding agents like Claude Code, GitHub Copilot CLI, Cursor, Gemini CLI, Codex CLI and other MCP clients, built on the iii engine. It stops agents from forgetting context between sessions, so developers don't have to re-explain the same project details repeatedly.

pydantic/pydantic-ai

Pydantic's typed agent loop covers voice, images, and every LLM — Built by the Pydantic team, this SDK offers a typed, extensible agent loop that works with any model via a simple string swap. The same agent runs across web, CLI, voice calls, and durable queues, aimed at Python developers wanting type safety across the whole AI stack.

datawhalechina/easy-vibe

A no-code-to-launch roadmap for Vibe Coding beginners — Easy-Vibe is a free, hands-on tutorial teaching non-developers to build real products with AI coding agents like GPT, Gemini and DeepSeek. It offers a full roadmap, step-by-step lessons and interactive demos in 10 languages, refined by feedback from tens of thousands of learners.

iii-hq/iii

A Rust engine for composing and watching services live — iii is a Rust-built engine with SDKs for JavaScript, Python and Rust that lets developers compose, extend and observe backend services in real time. It targets AI/agent and API builders who need visibility into running systems, distributed via Docker, npm, PyPI and crates.io.

googleapis/mcp-toolbox

Google's MCP server connects AI agents to enterprise databases — Renamed from Gen AI Toolbox, this Go-based MCP server lets Gemini CLI, Claude Code and other MCP clients query databases like BigQuery, PostgreSQL, Spanner and MongoDB via prebuilt tools, plus a framework for building secure custom NL2SQL and semantic-search tools for production agents.

kserve/kserve

CNCF platform unifying generative and predictive AI serving on K8s — KServe is a CNCF incubating project that unifies LLM and traditional ML model serving on Kubernetes. It adds vLLM/llm-d backends, OpenAI-compatible APIs, KV cache offloading and model caching for generative workloads, alongside multi-framework support, canary rollouts and explainability for predictive AI. Useful for MLOps teams running both LLMs and classic models at scale.

vshulcz/deja-vu

A local memory layer that mines your coding agents' own session logs — Deja-vu indexes months of session history already saved on disk by Claude Code, Codex, Cursor and 39 other agents, then surfaces the relevant fix to whichever agent runs next. It works via local search, MCP and hooks with no LLM calls, shipped as a single Go binary, and reports 97.2% R@5 on LongMemEval-S.