Frameworks & SDKs

CrewAI is a Python framework for building collaborative agents, role-based crews and controlled workflow flows.

LangGraph is an orchestration framework and runtime for building long-running, stateful agents and controlled agent workflows.

Mastra is a TypeScript framework for building AI agents and applications with tools, memory and workflow capabilities.

AutoGen is a programming framework for agentic applications and multi-agent workflows, retained for teams evaluating or maintaining existing implementations.

OpenAI Agents SDK provides application-level building blocks for agent workflows in Python and TypeScript.

Vercel AI SDK is an open-source TypeScript toolkit for building AI-powered applications and agents.

Agent Client Protocol standardizes communication between coding agents and client applications such as code editors.

Build and deploy custom remote MCP servers on Cloudflare Workers to expose tools, APIs, and Cloudflare services to AI agents like Claude and Cursor via standardized protocol.

The official C# SDK for the Model Context Protocol (MCP), maintained in collaboration with Microsoft. It enables .NET developers to easily build MCP servers and clients, exposing tools, resources, and prompts to AI agents like Claude, Cursor, and Gemini.

OpenViking is an open-source context database for AI agents that organizes memory, resources, and skills using a filesystem paradigm (viking:// URIs), enabling hierarchical context delivery, layered loading, and self-evolving agents.

Hindsight is an open-source agent memory system that enables AI agents to truly learn over time by retaining facts, recalling with hybrid strategies, and reflecting to form mental models.

Cloudflare's Dynamic Worker Loader API enables Workers to spawn new, isolated Workers at runtime with arbitrary code, providing lightweight sandboxing ideal for executing AI-generated code securely and efficiently.

The fast, Pythonic framework for building MCP (Model Context Protocol) servers and clients. Turn any Python function into production-ready LLM tools with zero boilerplate.

Microsoft A2A (Agent‑to‑Agent) is Microsoft’s implementation and support of the open Agent2Agent (A2A) protocol for interoperable AI agent communication across platforms and services.

Production-ready, easy-to-use multi-agent framework for building transparent, scalable LLM-powered applications with ReAct agents, tools, memory, evaluation, realtime voice, and secure runtime support.