Agent frameworks & infrastructure

Agent frameworks & SDKs

12,596 agents · most starred first

Agent frameworks & SDKs, page 34 of 126

concierge-os★ 9

Open Source AI platform to build Agentic services, ChatGPT Apps and MCP Servers

ai-maestro-plugins★ 9

Build complete AI agents — 150+ personalities + 7 skills + messaging + identity. The skill builder for AI Maestro.

claude-agent-sdk-skill★ 9

Production-ready guides for building AI agents with Claude Agent SDK in Python and TypeScript - complete coverage of official docs plus advanced patterns (subagents, hooks, batch processing)

bach★ 9

BACH v3.15.0: local-first text-based OS for LLM agents with SQLite memory, 113+ handlers, 550+ tools, MCP servers, scheduler and multi-agent orchestration

moe-sovereign★ 9

Self-hosted Compound AI System for sovereign environments. Features token-saving complexity routing, deterministic MCP tools for math/logic, configurable expert templates, causal/reinforcement learning, agentic loops, and a federated Neo4j GraphRAG.

reyn★ 9

AI agent workflow OS . Constrained, validated, replayable execution. Predictability over autonomy.

dify-aio★ 9

Dify for Unraid, bottled up as an AIO container: agents, RAG, workflows, plugins, sandboxing, Postgres/pgvector, Redis, and sane homelab defaults without the compose-file maze.

JavaChatX★ 9

Open-source AI Agent Development Framework built in Java - Build intelligent conversational agents with Gemini, and Free AI Models

neo★ 9

An open-source agent harness that makes small models do big work.

agentbox★ 9

Portable AI agent configuration packaging and sharing tool

custodian-labs-python★ 9

Build & host live AI agents with 5 lines of code.

agents-txt★ 9

agents.txt is a lightweight, machine-readable capability declaration layer for websites in the agentic web: a protocol-agnostic discovery file that publicly announces what agent-interaction capabilities a site supports, without embedding the implementation details of any specific protocol.

agentic_executables★ 9

Manage knowledge as apps | CLI | MCP | AI Knowledge Framework for AI Agents & Humans

ChatbotMakerspace★ 9

This repository provides a GitHub Codespace environment prepared to develop a transactional chatbot using Rasa Open Source for conversational flow and Node-RED for custom actions.

DIY-your-AI-agent★ 9

A minimal AI Agent runtime demonstrating tool calling, agent loops, context management, and extensible architecture.

eywa★ 9

🌿 Production-grade Go framework for conversational AI agents — LLM abstraction, RAG, memory, WhatsApp, scheduling & observability built in.

Amadeus-AI★ 9

Amadeus-AI is python framework for autonomous AI agents with planning, execution graphs, and plugins.

skillos★ 9

Agent SkillOS — the wealth-accumulation layer for self-improving AI agents.

agent★ 9

Domain-agnostic AI agent runtime (Zig + NAPI)

antigravity-ai-kit★ 9

Antigravity AI Kit is a Trust-Grade AI development framework with a 21-module runtime engine, 19 specialized agents, 31 commands, 31 skills, and 14 workflows — all backed by 261 tests and governance-first principles.

agent-harness-from-scratch★ 9

A minimal, production-shaped ReAct agent framework built from scratch in pure Python think-act loop, typed tools, layered memory, guardrails, and an eval harness. No LangChain.

cyntr★ 9

Enterprise AI agent platform. Self-hosted, open source, single binary. Policy engine, audit logging, multi-tenant, federation.

core-agent★ 9

Reusable Go base agent on the Google ADK — multi-provider (Gemini, Vertex, Claude on Anthropic + Vertex), MCP-native, with AGENTS.md, skills, and permissions wired up.

oh-my-fable★ 9

oh-my-fable — Fable 5's way of working a long task (plan first, self-correct, never lose the thread), as a model-agnostic agent harness. The run lives in one serializable RunContext, checkpointed every step, so a crash is a pause. Zero deps, deterministically testable.

demiurge-agent★ 9

Build file-backed, self-evolving Agent Cores.

Taroai★ 9

以 Agent Loop 为核心,统一 Skills、工具调用、沙箱执行与工作流自动化的开源 企业级 AI Agent 平台

bumblehive★ 9

An agent runtime designed for tool-driven loop engineering.

mainspring★ 9

An open-source operating system for long-lived, autonomous, revenue-generating agent businesses — memory, money, and governance for any LLM brain. Built and run by an autonomous AI agent.

cindy-protocol★ 9

Cindy 客户端与服务端共享的线上协议 · The wire protocol shared between the Cindy client and server

ts★ 9

Open agentic runtime in TypeScript — run any coding agent in the background, check its work before it lands, and project tools/skills/agents/workflows into CLI, HTTP, MCP, SDK, Mastra & the AI SDK.

Palamedes★ 9

An open-source goal discovery and synthesis engine that generates novel, detailed missions before planning begins.

Lingye_Agent★ 9

Modular Python framework for LLM agents with multiple agent patterns, memory, RAG, MCP, and context engineering.

graph-engineering★ 9

Zero-dependency Python graph runtime for agent loops · Mermaid + ASCII · LangGraph energy, bicycle size

nanoagentphp★ 9

Add AI agents to your PHP app in an afternoon - local-first, no Python sidecar, no framework to learn.

RE-Framework★ 9

Modular engineering methodology framework for AI agents — reverse engineering & software development (core + re-binary / re-code / swe modules).

ai-os-starter★ 9

The folder you talk to. A starter cockpit to build your own AI OS, built on the FRAME method. Run build-os and stack from there.

crewai-recipes★ 9

Ready-to-run CrewAI multi-agent automation recipes powered by NVIDIA NIM (Llama 3.1 8B by default, 70B optional). Clone, set one API key, run.

soulhub★ 9

SoulHub is an open-source Agent persona template community — share, discover, and deploy well-tuned AI Agents with one click.

mani★ 9

Declarative agent runtime in Go: define governable AI agents as YAML, not code — run them headless, as a service, or on triggers, with policy, limits and an audit trail built in.

omakase★ 9

Agents as plain Ruby objects. Fields are state, methods are tools, and the methods you declare without a body are written by the model at runtime.

alpine-agent★ 9

gradio region:us

golem★ 9

Go-first framework for dependable AI agents: typed dependencies and outputs, explicit tools, composable models, observable runs

genpark-dpll-sat-solver-boolean-satisfiability-skill★ 9

GenPark AI Agent Skill - DPLL (Davis-Putnam-Logemann-Loveland) Boolean satisfiability solver with unit clause propagation and pure literal elimination for agent policy verification.

genpark-first-order-logic-resolution-prover-skill★ 9

GenPark AI Agent Skill - First-Order Logic (FOL) resolution theorem prover applying Most General Unification (MGU) and refutation proof trees.

CrestApps.Core★ 9

CrestApps.Core is the AI application framework for .NET. It gives you the orchestration, management, chat, document, RAG, MCP, A2A, agent, reporting, and extensibility building blocks needed to add production-ready AI to an existing application without stitching together a pile of disconnected SDK samples.

dbos-vercel-ai★ 9

DBOS durable execution for the Vercel AI SDK

genpark-cocke-younger-kasami-cyk-pcfg-parser-skill★ 9

Autonomous Agent CYK (Cocke-Younger-Kasami) Probabilistic Context-Free Grammar (PCFG) Parser

genpark-earley-context-free-grammar-parser-skill★ 9

Autonomous Agent Earley Context-Free Grammar Chart Parser with Arbitrary Ambiguity & Recursion

talk-stop-prompting-start-planning-with-embabel★ 9

Talk materials for Stop Prompting, Start Planning with Embabel.

Ruvie-Agent★ 9

AI agent for company

agent-template★ 9

A Lego-style agent system template built from agent, MCP, CLI, and skills.

agent-blueprint★ 9

从 PRD/产品点子到可运行 Agent 产品的架构蓝图与参考实现,包含 17 个 Harness 组件、生产级脚手架、技能模板和测试。

idemkit★ 9

Make any operation safe to retry. Crash-safe idempotency for HTTP requests, queue messages, and function calls, in Python and TypeScript. Redis, Postgres, MongoDB, DynamoDB.

deepseek-r1-structured-outputs★ 9

This project demonstrates how to get structured outputs from Deepseek R1 reasoning model and do tool calling using an OpenAI summarization step. It's made using Next.js serverless routes and uses shadcn/ui for the frontend.

pipelineframework★ 9

Build with AI. Run with guarantees. TPF is a Java framework for strongly typed business flows that compose model decisions, authorised SaaS capabilities, and deterministic code—with durable effects, waits, retries, replay, observability, and flexible deployment.

AgentOne★ 9

开源企业级 AI Agent 中台:模型接入 / 知识库 RAG / ReAct Agent / Skill 生态 / MCP / IM 机器人,一键私有部署

jishushell★ 9

All your agents, one JishuShell

frankenst-ai★ 9

frankstate brings order and helps you build complex LLM workflows using scalable, testable, and reusable components.

dsh-plugin-jev-effort-selector★ 9

DeepSeek Harness (DSH) 推理等级自动选择插件:由 Jev System One 模型判断每条消息值多少思考量,按模型声明的等级自动推导档位,上下文信封让「继续」这类追问继承话题深度,低置信度向上取,任何失败都静默沿用原等级。 | Jev-driven reasoning effort per message: per-model ladders derived from what each model advertises, a fixed-size context envelope so follow-ups inherit topic depth, ties break upward, silent fallback on every failure path.

agent-sdk★ 9

TypeScript SDK for durable AI agents: tool calling, approvals, sessions, MCP, and a CLI. Runs in your own Node process.

k8s-ai-agent-platform★ 9

Private, self-hosted AI platform that runs on one laptop (no cloud, no GPU, no API keys): agentic RAG with citations, LangGraph + MCP tools, LiteLLM routing with an evaluation gate, MLOps (Dagster, MLflow, BentoML), OpenTelemetry tracing, SSO and Argo CD GitOps on k3s with Ollama.

self-evolving-agent★ 9

Turn a cron-triggered AI agent into a self-evolving digital employee — readonly SOUL + growable identity + heartbeat loops + hard-rule enforcement. Agent-native install: just send the repo link to your agent. 让定时任务 agent 变成会自我生长的数字员工

yai-agent-core★ 9

Embeddable, self-adaptive agent kernel for Python hosts: existing functions become tools, tasks auto-route (direct/react/plan/clarify), and every decision is an event. Kernel ships zero third-party deps; MCP, OpenAPI, browser and desktop are optional adapters.

mormot2-ai★ 9

mORMot2-based AI/LLM extension for Free Pascal and Delphi: MCP server, LLM clients, embeddings, vector stores, RAG

linkly★ 9

An LLM-native programming language: you declare intent, and it lowers to a semantic IR that runs on an interpreter or compiles through MLIR to a native binary.

NeuroCore★ 9

[ARCHIVED] A complete, self-hosted AI agent framework. Visual flow editor · Persistent memory · RAG · Tools · Multi-platform messaging · Community marketplace.

AgentStrata★ 9

A customizable agent runtime system for external applications, with LangGraph-based agents, application integrations, Harness and Eval.

agent-harness-engineering-pattern★ 9

agent-harness-engineering-pattern

python-ai★ 9

Durable execution for AI agents in Python: make LangGraph, CrewAI, Pydantic AI, OpenAI Agents, Google ADK, Strands and more fault-tolerant with Dapr Workflows on Diagrid Catalyst. Crash-proof agents, no re-run LLM calls.

xuen★ 9

随机灵感词 × 外部 agent 模拟研究者的 TUI 循环

self-evolving-agent★ 8

An OpenClaw skill that upgrades self-improving agents from reactive error logging to goal-driven capability evolution with curriculum, evaluation, transfer, and promotion.

openai-agent★ 8

Easy-to-use OpenAI Agent with support for the latest Function call feature.

mastra-app-template★ 8

mastra template with nextjs + assistant ui

freeact-skills★ 8

Predefined skills for the freeact agent library

gecko★ 8

Async-first, production-grade AI agent framework with workflow, RAG, observability & multi-agent teams(Python 智能体框架)

agentry★ 8

Compose and reuse your AI agents like React components

drive★ 8

🚀✨ DRIVE, the tool for creating and managing autonomous LLM agents; implemented using Apache Kafka, Docker, and LangChain.

semantic_kernel_copilot_demo★ 8

A console application demonstrating how to create a copilot using Semantic Kernel and its various features

linden★ 8

A Python engine for orchestrating complex LLM workflows, from simple data processing to RAG agents.

haive-tools★ 8

Tool implementations for Haive agents — search, calculators, APIs, and integrations

haive-core★ 8

Foundation for the Haive AI agent framework — engines, graph builder, state schemas, persistence, and stores

gai★ 8

🤖 GAI is a flexible Go framework for building agent-style applications on top of LLMs

Taskx-nano★ 8

A python library built to execute tasks with natural language

agentic-ai-patterns★ 8

Design patterns and reference implementations for autonomous AI agent systems. ReAct, tool-use, memory management, fallback, and multi-agent orchestration.

haive-agents★ 8

Production-ready agent implementations: SimpleAgent, ReactAgent, MultiAgent, MemoryAgent, RAG variants, and more

Taller-AgenteIA★ 8

Orion - Agente IA local con herramientas Este repositorio contiene el código completo y comentado de un proyecto educativo que muestra cómo construir tu primer agente inteligente con Python y Ollama.

ask-llm★ 8

The easiest way to supercharge your apps with LLM!

chat-bot-api★ 8

NestJs Ai chat bot api

tilores-langchain★ 8

This repository provides the building blocks for integrating LangChain, LangGraph, and the Tilores entity resolution system.

TypeChain★ 8

Proof of Concept Implementation of Langchain in Haskell

gpt-vector-agent★ 8

🧩 Interfacing with different LLMs-chains, vectorstore databases, and autonomous agents.

litechain★ 8

The zero-boilerplate LLM framework for humans.

ACE_Model_Implementation★ 8

A python implementation of Dave Shap's ACE Model

deepseek-langchain-groq-chatbot★ 8

An advanced AI chatbot implementation that combines the power of Groq's LLM with LangChain's tools for internet access, web searching, and conversation memory. This chatbot can search the internet, access Wikipedia, and maintain conversation context while providing informative responses.

fabricatio★ 8

Fabricatio is a Python library designed for building applications that leverage Large Language Models (LLMs) using an event-based agent structure. It utilizes Handlebars for templating and provides a robust framework for managing tasks, workflows, and toolboxes, enabling developers to create intelligent and efficient applications.

agents-template★ 8

A template repo for development with the AEA framework

anvil★ 8

Forge once. Run everywhere - universal tool compiler for AI agents.

multimind-sdk-js★ 8

JavaScript for MultiMindSDK – bring advanced context transfer and multi-LLM orchestration, Node.js apps, and web tools

konvo★ 8
agentic-demo-kotlin-spring-ai★ 8

A Kotlin Spring AI demo uses the Agentic approach to demonstrate "Let LLM think on graph schema"

More in Agent frameworks & infrastructure