Alternatives to Semantic Kernel

AutoGen is Microsoft's open-source framework for building conversational and event-driven multi-agent AI apps in Python and .NET.

Microsoft Agent Framework is an open-source SDK for building agentic AI agents and workflows in Python or C#, now with the GA Agent Harness

Microsoft's open-source, code-first Python framework for building stateful, data-intensive conversational AI agents.

Open source durable execution SDK for background jobs, workflow orchestration, and AI agents

OpenAI Agents SDK: a free Python framework for building multi-agent workflows with handoffs, guardrails, and sandbox agents

Hello Agents is a free, open-source 16-chapter tutorial for building AI agents from scratch in Python.

Hermes Agent: a self-hosted open-source AI agent with persistent memory and reusable skills across chat platforms.

TabTin is an open-source team agent harness where sub-agents and people share one task surface for code, docs and spreadsheets.

Open-source meta-harness that runs Claude Code, Codex and custom AI agents under one policy, sandbox and cost layer

Swarms is an open-source Python framework for multi-agent orchestration with named swarm topologies and reasoning agents

Open-source AI agent harness that plans tasks, runs tools, and keeps long-term memory across sessions.

Open-source multi-agent framework that assigns PM, architect, engineer and QA roles to GPTs for structured software tasks

AgentScope is an open-source Python framework for building multi-agent AI systems with debate, routing, and plan-based reasoning.

VCPToolBox is the open-source runtime that gives LLM agents persistent memory, tool calling and multi-agent coordination via the VCP

Open-source agent frameworks (langchain, langgraph, deepagents) plus LangSmith for tracing, evals, deployment and governance.

TypeScript agent framework for building, observing, and improving agents that run for days

Haystack is deepset's open-source Python framework for building inspectable RAG pipelines and production-grade AI agents

DeepAgents is an open-source agent harness for long-horizon, multi-step agents with sub-agents and a pluggable filesystem.

OpenAgentSkill is a task-to-skill resolution API that turns a natural-language job into one ranked, install-ready agent skill with trust

Open-source AI agent orchestration: run a team of AI agents with org charts, budgets, and approval gates.

Designing Multiagent Systems: a Python book on multi-agent LLM architecture, MCP, and picoagents

Xagent is an open-source AI agent engine that plans tasks from stated outcomes instead of hand-drawn workflows.

Free, open-source desktop GUI that runs the self-hosted Hermes Agent with a closed learning loop that writes its own skills

Learn Claude Code is an open-source tutorial that builds a Claude Code-style agent harness in Bash across 20 runnable sessions.

OpenAgents runs Data Agent, Plugins Agent and Web Agent from one open chat UI, self-hostable under Apache-2.0.

Agenta is an open-source workspace for building and automating AI coworkers that work in Slack or Telegram

Trigger.dev runs durable AI agents and long-running background jobs in TypeScript with no execution timeout.

SmolAgents builds code-writing AI agents in Python — agents that think and act in code, sandboxed and model-agnostic.

Clawdi hosts always-on Hermes and OpenClaw AI agents in the cloud with shared memory, skills, and vaults across every agent you run.

Open Multi-Agent is an MIT-licensed TypeScript runtime for audit-ready multi-agent orchestration with durable approvals.