Agent frameworks & infrastructure
Agent frameworks & SDKs
12,596 agents · most starred first
Agent frameworks & SDKs, page 26 of 126
RiveScript is a scripting language for authoring chatbots.
The LLM-powered function builder for TypeScript
Create interactive, powerful AI applications with streaming, structured responses from language models
opengpt is a open implementation of GPT agents.
A Python module that lets you whip up your own custom chatbot with just a few lines of code. This AI-based package feeds your custom data into a neural network which is trained and outputs a fully functional chatbot.
Multi-Node LLM Processing Framework
Haive AI Agent Framework — production-ready monorepo with 53+ agents, RAG, memory, KG, and games
Protocol-only AI agent framework — bring your own everything
Open source implementation of Anthropic's Programmatic Tool Calling
TSCG - Deterministic tool-schema compilation for LLM agents, including text-mode tool use without native tool calling and context-budget RAG. Research: arXiv:2605.04107 and 2605.26165. TypeScript, MIT.
AgentBrook
🚀 Production-ready AI agents framework. Fork → Deploy → Ship in minutes. Multi-agent patterns, FastAPI backend, observability with Opik. Built with Google ADK or Langgraph. MIT License.
Meerkat - an operational agent platform.
Minimal Go agent framework for LLM tool-calling loops (v1.0, stable). Typed tools, parallel execution, three hooks, streaming progress, snapshot resume, prompt caching across iterations. Anthropic Claude, OpenAI GPT-5, Google Gemini via pi-llm-go.
Local-first, extensible AI agent workspace — FastAPI + Next.js + LangGraph, with skills, MCP tools, artifacts, long-term memory, sub-agents, and multi-provider (DeepSeek/OpenAI/Anthropic) reasoning streaming.
This is a framework intended to help developers build applications in VBA that interact with OpenAI's large language models like GPT4, DALL-E and ChatGPT
AI agent with zero dependencies, bring your own bun
Run a fleet of AI coding agents like a team: seats, messages and exact-commit reviews. Local-first, one static binary.
Declare the outcome, skip the logic. The developer-first infrastructure engine for structured, multi-turn AI conversations with built-in safety, async follow-ups, and native compliance
The Mozaiks core is a self-hostable runtime for deploying AI-driven, multi-tenant web apps with real-time UI, auth, and billing built in. Apps are defined via YAML and lightweight Python/JS stubs for agent logic, workflows, and integrations — so developers focus on behavior, not plumbing. :link: https://docs.mozaiks.ai/ :link: https://mozaiks.ai
An outcome-first operating framework for reliable AI agent work.
Local-first, model-agnostic LLM agent-harness research. Restart planned after a hiatus; existing code and historical claims await revalidation. MIT.
A state-based autonomous agent framework. Crafting digital entities with non-linear cognitive memory and self-evolving emotions. (Work in Progress)
Let an LLM write its own tools at runtime: parsed with acorn, versioned in SQLite, run in a supervised worker process
Starter kit to deploy LLM agents with memory and RAG using LangChain, OpenAI, and Google Cloud (Cloud Run, BigQuery).
The batch engine for AI agents.
A decision layer embedded into the harness tool-selection loop
A lightweight, embeddable, data-centric scripting language whose reach extends from JSON/FD data all the way down to runtime-compiled C and raw machine code.
The framework that doesn't get in the way of you or your agents.
An opinionated development framework for building production-ready AI agents with LangGraph. It grounds AI coding assistants (Cursor, Windsurf, Cline) and guides them to use local, official documentation, ensuring reliable, secure, and observable agentic workflows.
Build your own AI-powered automation tools in the terminal with this extensible agent framework
🧙♂️ Zero-Code Machine Learning Wizard - Transform ideas into intelligent solutions without writing code. AI-powered ML pipeline automation with interactive web interface.
Open-source LLM agents framework for token systems simulations — pioneered domain-specific cognitive architectures with plan-and-act agents for data science process automation, built on cadCAD/radCAD
Part of the LangChain.NET project. Has separate abstractions, does not contain dependencies on the main project and can be used independently
An open-source, code-first C# (Dotnet) toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
Open-source alternative to OpenAI AgentKit — build, manage, and deploy AI agents with full transparency and flexibility.
Strongly-typed LLM Function Calling examples, run on OpenAI, Ollama, Mistral and others.
Unite and orchestrate AI agents. A production-ready ADK for building full-stack, agentic AI solutions with multi-provider support.
Unified API for LLM providers and simple agent library in JS, Rust, and Go
Blue is an open-source framework for building enterprise-ready agentic workflows through compound AI system architecture. Blue uses streams for orchestration, structured communication for control, and seamless integration with existing infrastructure.
Minimal, async-first Python framework for production LLM apps- 2 hard deps, no magic, no SaaS.
The open-source client-agnostic agent harness
TypeScript AI agent framework where personality is architecture. Specialists, not chatbots.
文 Veyllo Agentic Framework
AladdinAI: open-source BYOI (Bring Your Own Infrastructure) AI Agent platform. Connect your own VMs, NVIDIA NIM, OpenAI, Anthropic, Ollama — build and manage AI agents visually.
Intent Agent Model (IAM) — Bob reference runtime on Google ADK. Historical V1. IAM = Intent Agent Model, not Identity and Access Management.
Agents that talk through model internals — activations & J-space — instead of text. No words pass between them. The lab on top of brainscope + hidden-directions.
Open-source alternative to LangGraph Platform for TypeScript — self-host your LangGraph.js agents behind the standard Agent Protocol API. Your infra, your Postgres, no vendor lock-in. Drop-in for the LangGraph CLI.
Open-source multimodal AI framework & agent harness — an enterprise AI runtime for building agentic applications over video, audio, images, documents, and organizational knowledge. MCP tools, A2A orchestration, persistent memory, hybrid RAG, GraphRAG, evaluation, guardrails, model lifecycle.
Multi-tenant projects & users plugin for DeepSeek Harness (DSH): per-user symlinked workspaces, login gate, session isolation, workspace-write lock and system-prompt guard
Agentic coding framework and meta-harness for planning, building, reviewing, and tuning software across coding agents. Portable agent squads, skills, governed project knowledge, quality gates, and local token observability. Human judgment. Agent speed.
给它一个 PRD、一个需求点子、或一个可落地的 Agent 方向,它推导出架构方案、组件选型与代码骨架,直到一个能在真实环境运行的产品。
⚡ Fastest and lightweight harness for AI agents
A bunch of LLaMa model investigations, including recreating generative agents (from the paper Generative Agents: Interactive Simulacra of Human Behavior)
Platform and API Agnostic library for powering chatbots
SAP Conversational AI official SDK for Swift
AI Chatbot Starter Kit: An open-source, extensible framework for rapidly developing custom AI chatbots with integrations for popular data sources, messaging platforms, LLM models, and CRM systems. Ideal for developers looking for a minimal boilerplate solution.
LangChain: Rapidly Building Advanced NLP Projects with OpenAI and Multion, facilitating modular abstraction in chatbot and language model creation
Simple Docker wrapper for LangFlow
Rasa Docker image for ARMv7. Runs on a Raspberry Pi.
A neural network-based AI chatbot has been designed that uses LSTM as its training model for both encoding and decoding. The chatbot works like an open domain chatbot that can answer day-to-day questions involved in human conversations. Words embeddings are the most important part of designing a neural network-based chatbot. Glove Word Embedding and Skip-Gram models have been used for this task.
chatterbot (machine learning bot) and Wit.ai in Flask with websocket
简单、易用、稳定、高效,便于扩展和集成的,大语言模型工程化开发框架
⚡️ Asynchronous framework for ChatGPT API 🤖
ChatGPT plugin example using Cloudflare Workers and Clerk OAuth2 backend.
OpenAI devday example 🗺
BMAD AI/ML Engineering Expansion Pack - Streamlined framework for AI Singapore programs (MVP, POC, SIP, LADP) with specialized agents, workflows, and templates for ML/LLM development
Type-safe AI agents for Go. Suricata combines LLM intelligence with Go’s strong typing, declarative YAML specs, and code generation to build safe, maintainable, and production-ready AI agents.
Sendbird AI-Agent
LLMs based multi-model framework for building AI apps.
Durability for web streams powered by S2
SAP Conversational AI official SDK for PHP
⚙️ List of Crew-Ai 👨👨👦👦 Tools 🛠️ (Webscrapper, PDF-RAG, Composio, YouTube-RAG...)
Autonomous AI Agent for the JVM: shell, files, google search, runs any LLM generated java code on the jvm itself. JIT compilation and a child-first classloader with ANY classpath. Comes wit a swing GUI , can be "dropped in" into any existing java application. Easy tools easy pojos. IoT devices. Run it standalone for a pure-java AI assistant.
An Open Source Typescript implementation of Programmatic Tool Calling for AI Agents (Based on Code Mode)
Visual AI agent builder. Drag, drop, and generate LangChain/LangGraph code using natural language or a node-based UI.
Intelligent AI Chatbot built using Java Spring Boot and Spring AI. Engages in dynamic conversations leveraging OpenAI models.
Django AI Core provides developer-focused AI features for implementing AI tooling in to Django sites.
Open-source platform for reproducible AI/ML workflows, from local experimentation to distributed workers and long-lived services.
A full-stack AI assistant built on Amazon Bedrock AgentCore. This sample demonstrates how to build a feature-rich, self-hosted AI assistant that leverages AgentCore capabilities like managed runtimes, sandboxed code execution, and persistent memory.
🐉 Dante Cloud 是支持阻塞式与响应式服务并行的微服务云原生基座和面向AI Agent开发的智能基座,基于领域驱动模型(DDD)设计,以高质量代码与低安全漏洞为核心,高度模块化、组件化,覆盖IoT设备认证、国家等保三级、接口国密数字信封加密等多层次安全体系,并提供多租户解决方案。独创“一套代码实现微服务和单体架构灵活切换”,赋能企业级智能应用构建。🔝🔝点个star 持续关注更新!
Self-extending AI agent system running on local Ollama - creates its own capabilities autonomously
TypeScript AI agent framework with recorded context, tool calls and decisions, plus configurable answer checks.
Spring Boot + Spring AI 实践项目:聊天、RAG、工具调用、MCP、记忆、SSE 与 Vue 前端联调。
Open-source full-stack AI Agent workbench—streaming chat, visual workflows, RAG eval, multi-model routing, and observable context & tool calling.
A Blazor Web App playground for quickly testing agents built with Microsoft Agent Framework
A state-based autonomous agent framework. Crafting digital entities with non-linear cognitive memory and self-evolving emotions. (Work in Progress)
Open-source full-stack AI agent with an embeddable React UI and FastAPI runtime. Self-host with streaming, pause/resume, durable state, and parallel multi-agent subtasks.
Local-first CLI, package contracts, and execution evidence for portable AI employees across Agent Hosts.
One toolkit for a team of AIs on different models.
Open-source System One decision engine. Typed choice / score / noul with calibrated probabilities. Self-host with Ollama or any OpenAI-compatible model. Apache-2.0.
One Rust UI declaration for SwiftUI, WinUI 3, GTK 4, Jetpack Compose, GPUI and the Web — desktop, mobile, watch and TV. Compile-time checks, co-located NativeView, and Skills + CLI for AI agents.
Local-first multi-agent AI workbench. Offline installers for Windows and Kylin/Linux (x86_64 and ARM64), with Skills, MCP, and a browser sandbox. Your data stays on your machine.
Interactive playground to experiment with Microsoft's Semantic Kernel SDK. Built with Next.js and .NET 7.
v0.1.0-alpha
A collection of Python agent samples built with the Google Agent Development Kit (ADK), demonstrating integrations with services like BigQuery and Vertex AI Search.
AI Agent Tools library for Graphlit Platform
LLM as Agent
Turn GraphQL APIs into LLM Tools
This project is an Express.js server integrated with OpenAI's API. It includes features such as file uploading, creating and managing an OpenAI assistant, and handling chat interactions using the assistant.