New in week 43

Framework for building, running and evaluating autonomous agents, with a planner-based multi-agent mode, skills and agent RL training.

Minimal Python graph framework of about 100 lines for LLM agents, workflows, RAG and map-reduce pipelines.

TypeScript agent framework on a Rust runtime, where runs are checkpointed, replayed with zero LLM calls and resumed after a crash.

Go implementation of LangChain for writing LLM-based programs from composable building blocks.

TypeScript platform to build and deploy long-running AI agents and workflows with retries, queues and human approval steps.

Go framework for multi-agent systems with graph workflows, tools, session memory, MCP, A2A, AG-UI and OpenTelemetry.

Open-source agent harness that runs the model loop for you, with MCP tools, skills, sandboxing, approvals and a chat UI.

Open agent harness for AI applications, with a YAML-configured coding agent, sessions, tools, MCP support and a CLI.

Python and TypeScript SDK that lets an agent message people on email, Slack, Discord, Telegram and other channels through one handler.

Python toolkit for Vietnamese NLP with a multi-provider agent module that calls LLM APIs using only the standard library.

TypeScript library that lets agents call MCP and UTCP tools by running model-written code in a sandbox.

Official Python implementation of the Universal Tool Calling Protocol, calling HTTP, CLI, MCP and text-file tools through plugins.

Python framework for autonomous agents that work in a sandboxed workspace, and task agents with tools and MCP support.

Python library that builds a tree index of long documents and has an LLM reason through it to retrieve answers, without a vector database or chunking.

TypeScript toolkit for AI apps and agents with a unified provider API, streaming, UI hooks and a wrapper for running established agent harnesses.

Workflow SDK: Build durable, reliable, and observable apps and AI Agents in TypeScript

Python toolkit for signed identities, verifiable credentials and delegated authority for AI agents, with MCP and REST interfaces.

Python SDK for building agents on Volcengine AgentKit, with a bundled Web UI, Studio deployment and short- and long-term memory.

TypeScript framework for AI agents with tools, memory, RAG, MCP and sub-agents, paired with a Cloud or self-hosted console.

Python framework for multi-agent applications where agents with defined roles talk through explicit flows, built on the OpenAI Agents SDK.

Python agent framework for OpenAI-compatible agents with tools, memory, MCP connections, multi-agent collaboration and workflows.

Event-sourced Python graph runtime where agent behaviors react to a shared graph, with an audit trail and fork-and-diff replay.

Open-source LLM function calling framework for AI agents, with serverless LLM tools and geo-distributed infrastructure.

Open infrastructure for discovering, evaluating, creating and routing reusable AI agent skills, with a Python SDK.

Managed Microsoft Foundry service for building, deploying and scaling AI agents, with a Python client library on PyPI.