Home › Directory › Agent Frameworks

Agent Frameworks

Libraries and platforms for building, orchestrating and running agents.

JSON: /api/v1/directory.json (filter on category=agent-frameworks)

Library for building stateful, multi-step agents as graphs, with persistence and human-in-the-loop controls.

Widely used toolkit of components for LLM apps, covering models, tools, retrievers and chains.

Framework for orchestrating teams of role-playing agents that split up a task.

Microsoft's framework for multi-agent conversations and workflows, in Python and .NET.

Lightweight SDK with agents, handoffs, guardrails and tracing as its core primitives.

Framework centred on connecting LLMs to your data, with ingestion, indexing and agent workflows.

Type-safe Python agent framework from the Pydantic team, with structured outputs and dependency injection.

TypeScript framework for agents, workflows, memory and evals.

TypeScript toolkit for streaming text, tool calls and agents across many model providers.

Minimal library where agents write and run Python code to complete tasks.

Microsoft's SDK for adding models, plugins and agents to C#, Python and Java apps.

Google's open-source toolkit for building, evaluating and deploying agents and multi-agent systems.

Build stateful agents on Workers and Durable Objects with scheduling, websockets and MCP support.

Programming framework that compiles prompts and weights from code and metrics rather than hand-written prompts.

Visual workflow automation with AI agent nodes; source-available and self-hostable.