Open-source framework for building multi-agent systems in Python or TypeScript, with rule-enforcing agents, workflows, MCP and A2A support, hosted by the Linux Foundation.
CATEGORY
SDK & Decision Frameworks
Build AI applications with structured decisions and agent workflows. Explore SDKs and frameworks for connecting models, tools, and application logic.
- products
- 38
- related topics
- 3
Explore products38
Showing 1–24 of 38 projects
Open-source multi-agent framework for Python and JavaScript that builds agents, agent teams and flows with memory, RAG, MCP tools and 100+ LLM providers, and can run them as chat bots.
An open-source Python framework for building local RAG and agent applications with small, specialized models that run on laptops and private servers.
A conversational agent framework that selects relevant guidelines, tools, domain knowledge and journey steps for each turn of a customer interaction.
A guardrail framework that combines prompt-injection detection, agent-alignment checks and generated-code scanning for LLM applications.
An open-source Python library for adding programmable input, output and conversation guardrails to applications built around language models.
A TypeScript framework for AI agents and apps with memory, tools, MCP and observability.
An open-source Python and TypeScript SDK for running AI agents with tools, structured output, sessions and multiple model providers.
A multi-agent platform and framework for orchestrating teams of AI agents.
A Python framework for extracting, classifying and generating structured data with language models, and coordinating agents through tasks and conversation threads.
An open-source agent SDK plus AgentOS, a runtime for building, deploying and improving agents in your own cloud.
A Python library for typed LLM calls, output validation and automatic retries based on explicit requirements.
A Python framework for programming rather than prompting language models, with algorithms to optimize prompts and weights.
A Python framework built on the OpenAI Agents SDK for coordinating agents with distinct roles, validated tools and explicit communication flows.
LangChain's open-source framework for building controllable agents as graphs, with durable execution, memory, streaming and human-in-the-loop steps.
An open-source library from .txt that constrains LLM generation so outputs always match a JSON schema, regular expression or grammar.
An open-source TypeScript agent framework for connecting models, tools, memory, retrieval and workflows, with observability and evaluation integrations.
An open-source TypeScript agent framework and product stack with a runtime, applications, and plugins.
An open-source library for getting validated, structured outputs from LLMs, with versions for Python, TypeScript, Go and Ruby.
Google’s open-source framework for building AI applications with model integrations, structured output, tool calling and a local debugging interface.
TypeSafe AI’s model for answering typed decision questions with structured values and probability distributions.
OpenAI's lightweight open-source framework for building multi-agent workflows with tools, handoffs, guardrails and tracing.
A framework for building AI agents and graph-based multi-agent workflows in Python and .NET, with streaming, checkpointing and human-in-the-loop controls; a Go SDK is maintained separately.
A TypeScript toolkit for building AI applications with model providers, streaming responses, and frontend integrations.