175 listed
Harnesses & SDKs
Frameworks and SDKs for building agents: the Claude Agent SDK, OpenAI Agents SDK, LangGraph, CrewAI and many more. Each page lists languages, packages, MCP and A2A support and how it orchestrates agents, with a link to the page every fact came from.
Showing 1–9 of 9
- 0.777score
ADK for TypeScript
Google
TypeScript toolkit for agents with Zod-typed tools, graph and multi-agent workflows, and Node.js or browser runtimes.
HarnessAFreshness: A★ 1.4K↓ 143.5K/wk - 0.745score
BeeAI Framework
IBM / Linux Foundation
Python and TypeScript framework for multi-agent systems, with MCP and A2A interop, YAML orchestration and OpenTelemetry tracing.
HarnessAFreshness: A★ 3.4K↓ 5.4K/wk - 0.740score
Deep Agents
LangChain
Opinionated agent harness on LangGraph with sub-agents, filesystem access, context management, persistent memory and approval of tool calls.
HarnessAFreshness: A★ 30.6K↓ 1.9M/wk - 0.737score
Mastra
Mastra
TypeScript framework for AI agents and apps with graph-based workflows, memory, suspend-and-resume approvals, MCP servers and model routing.
HarnessAFreshness: A★ 29.1K↓ 1.4M/wk - 0.734score
Google Agent Development Kit (ADK)
Google
Open-source toolkit for building, evaluating and deploying agents in Python, TypeScript, Go, Java and Kotlin, tuned for Gemini but model-agnostic.
HarnessAFreshness: A★ 22K↓ 2.4M/wk - 0.729score
Agent Squad
2FastLabs
Open-source framework that routes each user query to one of several AI agents, with Python, TypeScript and Swift runtimes.
HarnessAFreshness: A★ 7.8K↓ 1.5K/wk - 0.699score
VoltAgent
VoltAgent
TypeScript framework for AI agents with tools, memory, RAG, MCP and sub-agents, paired with a Cloud or self-hosted console.
HarnessAFreshness: A★ 10.9K - 0.659score
OpenAI Agents SDK (JavaScript)
OpenAI
TypeScript and JavaScript framework for multi-agent workflows with tools, handoffs, guardrails, sandboxes and voice agents.
HarnessAFreshness: A★ 3.9K↓ 1.2M/wk - 0.581score
code-mode
universal-tool-calling-protocol
TypeScript library that lets agents call MCP and UTCP tools by running model-written code in a sandbox.
HarnessBFreshness: B★ 1.6K