Harnesses & SDKs

Open-source Python toolkit for RAG and agents over your own data, with event-driven workflows and 300+ integration packages.

Open-source Python and TypeScript SDK for model-driven agents, with tools, MCP, multi-agent patterns and streaming.

Python SDK that runs the Claude Code agent loop inside your own program, with custom tools and hooks written as Python functions.

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.

Open-source Python framework for multi-agent systems: agent societies, memory, RAG, code interpreters, synthetic data and large-scale simulation.

CLI-first Python agent framework and coding agent with MCP, A2A and Agent Skills support, live streaming and human input.

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

Lightweight Python framework and starting template for agents that let the model choose its own approach to each task.

Policy enforcement, identity, sandboxing and reliability tooling for autonomous AI agents, installable with one pip command.

Python and TypeScript SDK that gives an AI agent a phone number, with pluggable LLM, speech and telephony providers.

Open-source reinforcement learning framework that trains LLM agents on multi-step tasks using GRPO.

Python framework that models an application as a state machine, with a trace UI and persisted state.

Python framework for LLM apps built from collaborating agents that hold conversation state, tools and vector-store memory.

Python library that turns files in S3, GCS and Azure into versioned, typed datasets for pipelines and agents.

Python agent runtime with an SDK for building your own agent, and an app that adds plugins and channel integrations.

Python library for autonomous agents that run scheduled tasks and exchange messages on the Fetch.ai network.

Python framework for real-time voice and vision agents that process video with pluggable models and processors over WebRTC.

Python runtime and framework for long-running agents, with durable sessions, approvals, spend caps and multi-agent collaboration.

Open-source infrastructure for continually self-improving agents that serves requests, learns from feedback and versions updates.

Reliability layer for tool calling with self-hosted LLMs, offered as a proxy, a workflow runner and middleware.

Open-source Python evaluation framework with ready-made and custom graders for scoring AI agent quality and producing reward signals.

Python library of 35 agentic AI architectures with one run() interface, built on LangGraph and multiple LLM providers.

Python multi-agent framework with PEER and DOE collaboration patterns and configurable LLM vendors such as Qwen, DeepSeek and OpenAI.

Python framework for durable AI agents built as dataflows that resume after crashes, retry safely and record each step.