Python agents

[ICLR 2026 Oral] SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM Agents

Agentifying Patient Dynamics within LLMs through Interacting with Clinical World Model

MolClaw: An Autonomous Agent with Hierarchical Skills for Drug Molecule Evaluation, Screening, and Optimization

50+ curated LLM observability tools PLUS 26 Agent Skills (several with runnable, unit-tested scripts) to build, evaluate, debug, secure & monitor reliable LLM apps. Tracing, evals, guardrails, LLMOps.

LangChain + LangGraph + LangSmith 入门教程实践

AI agent fleet orchestration terminal — bidirectional socket IPC, resource governance, mission-control dashboard. macOS · Windows. Built largely by AI agents under human direction.

gradio region:us

gradio region:us

gguf abliterated uncensored text-generation base_model:huihui-ai/Huihui-Qwen3-Coder-Next-abliterated base_model:quantized:huihui-ai/Huihui-Qwen3-Coder-Next-abli

transformers safetensors qwen3_5_moe image-text-to-text text-generation-inference unsloth

MCP server serving Apple Human Interface Guidelines as structured design tokens for AI coding agents — post-WWDC25 system colors, Liquid Glass constraints, SwiftUI mappings.

Engineering infrastructure for building retrieval, memory, and context-aware AI systems.

AgentOS: Serve agents over API, MCP, and chat interfaces like Slack

Aureon OS — Harmonic Nexus Core: a grounded AI operating layer for evidence-heavy, high-control work. Autonomous operator with a conscience in the loop, trading research, planetary/HNC research, and a self-building coding organism — one auditable system. By R&A Consulting (Aureon Zorza Technologies).

Autonomous research loop skill for LLM agents — inspired by Karpathy's autoresearch. Define a goal, let the agent iterate.

Official code for the paper: "Multi-User Large Language Model Agents"

CADSmith (formerly AutoFab) is a multi-agent system that generates manufacturing-ready CAD models from natural language by combining LLM code generation with programmatic geometric validation from the OpenCASCADE kernel and visual inspection from an independent vision-language model, iteratively refining until the part is dimensionally correct.

AI 全链路音乐创作 SaaS — MiniMax API + FastAPI + React，支持 Agent 工具调用 / 歌词 / 音乐 / 翻唱 / 语音合成 / 声音克隆，带 docker compose 一键部署

AI Reputation Manager for Claude Code. Analyze online reviews, score brand sentiment, benchmark against competitors, generate review responses, create crisis playbooks, and produce PDF reputation reports — 14 skills, 5 parallel agents.

Workspace for documents with AI

MCP server for SAP GUI for Windows automation via the SAP GUI Scripting API.

会读空气的本地 AI 伴侣核心模块——硅基感知 / 漫想模式 / 工具调用调度器(分批开源中)

Fast, deterministic document parser and chunker for RAG and agent pipelines. Ships an MCP server for agents navigating long documents with precision.

UbD 逆向设计驱动的教材写作 skill 组合，一键装进 Claude Code / Codex / WorkBuddy：教学定位 → 五件套 gate → 章节树 gate → 四段式逐章写作 → 审核定稿。学科无关内核 + 可插拔学科档案，理科真算复核、人文社科逐字核对引文出处，验证不了就标注「需作者确认」；关键决策作者拍板，中断随时续写

AI-платформа подготовки учебных материалов для преподавателей и технических спикеров: агент с долгосрочной памятью превращает ваш авторский материал в лекции, конспекты, методички и слайды. Self-hosted, любые OpenAI-совместимые модели.

Open-source, self-hostable delivery platform for Claude Code style agents. Connect Slack today, with more channels next. Run the same bundle locally and on Kubernetes, and ship it with git push.

Repo blueprint para Claude Code: no trae la app, trae las instrucciones para construirla en tu máquina paso a paso. Un agente que atiende cada chat de WhatsApp como setter y closer: califica, responde la objeción, agenda y escribe el CRM. 16 archivos por fase, 30 versiones fijadas, compuerta de 23 chequeos. En español.

Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents

An End-to-End Multimodal ECG Interpretation Agent. This agent solves the challenges of non-direct analyzability of ECG images and insufficient association of professional diagnostic knowledge. By adopting multimodal AI and knowledge-enhanced retrieval techniques, it generates accurate and implementable intelligent ECG diagnosis reports.

OpenAI compatible API on top of the free web clients

A terminal-based, multi-domain support-ticket triage agent grounded in a local support corpus.

An extensive agent playbook for automated porting of Python to Rust

Built-in memory decides what to remember. This one asks. Agent memory as plain files in your folder: the agent proposes, you approve, every line dated. Injected on Claude Code, Cursor & OpenCode; skills on Codex & Copilot. One memory per client, session handoffs, promotion into knowledge and rules on your yes, plus orchestration and QA layers.

Java Web 全链路自动化安全审计系统。基于 Claude Code Agent Teams，一行命令完成路由识别→调用链追踪→8种漏洞检测→利用链编排→审计报告。覆盖 SQL 注入/反序列化/SSRF/表达式注入/XXE/文件上传/文件读取/鉴权绕过 + OSV.dev 3000+ CVE。支持快速模式、增量审计、CFR 反编译。

Durable local inference for Oh My Pi: NInfer + Qwen3.8 27B on one RTX 3090/4090/5090, with restart-resumable OpenAI Responses state.

Yandex Wordstat API skill for AI agents — collect RU search semantics, frequency, and trends automatically.

Data science and Machine Learning AI agent

Open-source AI agents for coding, research and automation. Run locally with your models, persistent memory, multi-agent execution, skills, MCP and self-evolving.

A lightweight local mock LLM provider — deterministic OpenAI and Anthropic responses, with no API calls, no tokens, and no code changes.

A research terminal for prediction markets, with multi-agent evidence gathering, probability maps, and inspectable answers.

The open infrastructure standard and reference implementation for AI agent cooperation, discovery, and marketplace.

A macOS menu-bar app for voice control of Claude Code and Codex. When a session stops, it calls you, and you answer by voice or on Telegram. The call uses OpenAI's GPT-Live voice model on the ChatGPT account you signed Codex into, so you don't need an OpenAI API key.

A simple, local-first RAG framework for building document Q&A applications

Nexa - The FIRST-EVER Harness Native Agent Language

This repository focuses on the collaborative development of a snake game using the AutoGen framework. Leveraging multiple agents, each with a distinct role, the project aims to design, implement, and test a snake game that is both entertaining and challenging.

Action Engine for AI Agents

[AAAI 2026]RAG, Benchmark, robust RAG generation

Trying to extend AI Agents to a full topology of devices using CML

App built with lang-chain utilizing OpenAI API to generate cool pet names based on the animal type and color.

AI for Network Troubleshooting: A Proof of Concept with OpenAI and Langchain

AI Chatbot that helps you learn how to use Dagger

Multi-Agent LLM Evaluation Docs: https://maseval.readthedocs.io/

FAQ's answering chatbot using open source chatbot framework Rasa Stack

Design and build a chatbot using data from the Cornell Movie Dialogues corpus, using Keras

LLM-based chatbot capable of interfacing with external systems for knowledge retrieval and command execution

使用QQ聊天记录制作属于你自己的聊天机器人

GPT-For-NLG-Tasks

Python3 wrapper/library that simplifies telegram bot development. It supports both sync and asyncio methods of programming. Dive in and start making awesome bots!

A chatbot that can be used by businesses to communicate with customers via Whatsapp SMS using the Twilio API and OpenAI's GPT-3 language model.

Pytorch Implementation of "Sequential Dialogue Context Modeling for Spoken Language Understanding( https://arxiv.org/abs/1705.03455 )"

Dolores is a Python library designed to improve the developer experience when working with pretrained language models. Dolores provides prompts for interacting with language models that result in interesting or useful outputs.

Prefect integrations for working with OpenAI.

quick way to build a private large language model server and provide OpenAI-compatible interfaces | 快速搭建私有大语言模型（LLM）服务，提供OpenAI兼容接口

This repository helps you evaluate your models on the FreshStack benchmark!

AI web agent to find answers to any question

智能派单Agent系统是一个基于大语言模型(LLM)和LangChain框架的智能对话系统，用于为用户提供自动化的技师派单服务。系统具备上下文记忆能力，能够理解用户的多轮对话，并根据用户需求匹配合适的技师资源。

LTL Optimal Multi-Agent Planner (LOMAP)

Turn any MCP Client into a "multi-agent" system (via prompting)

This repository is your comprehensive guide to building, orchestrating, and deploying AI agents and workflows using the Microsoft Agent Framework. Whether you're creating simple chat agents or complex multi-agent workflows, this repository has everything you need to get started.

Codes accompanying the paper "Influence-Based Multi-Agent Exploration" (ICLR 2020 spotlight)

Multi-Agent RL Environment for the Stratego Board Game (and variants)

Basic ReAct agent implementation in Python from scratch

A platform designed to facilitate the development of advanced conversational agents using retrieval augmented generation (RAG).

Opus Agents is an open-source Agentic AI framework, that helps you build AI Agents and Tools that run reliably using abstractions like Custom tool, Higher-order tool, Meta tool. It comes bundled with Agents like OpusTodoAgent that work with Productivity & Collaboration software

replicantlife is a framework for generative agents that can be used in a simulation engine or standalone. Agents are powered with metacognition modules that allow that to learn and adjust their strategy over time.

Paused Codex App auth-split proxy experiment; future work would need a Responses-to-Chat-Completions adapter.

Streamlit web app utilizing OpenAI (GPT-4) and LangChain LLM tools. Application includes an SQLite DB for login/authentication and message storage for later retrieval. Users can upload/embed their own PDF documents for chatbot reference. The user can then interact with a GPT-4 chatbot intended for the user's specified input, context, and scenario.

A YouTube agent built using AutoGen

AI code generation and improvement

Echo is your personal AI assistant on Telegram, designed to enhance your productivity by providing a seamless needs. With MongoDB integration, your reminders, schedules, broadcast, APIs and others persist even if the bot restarts, ensuring you never miss a moment.

AI assistant, based on the GPT-3.5 model by OpenAI, designed to enhance your proficiency in writing research papers. Allows you to adapt your content to academic standards, transform bullet points into eloquent text, or enhance the quality of your writing through error detection.

Kel - CLI-based AI assistant

ZERG (Zero-Effort Rapid Growth): Secure and parallel Claude Code execution system. Overwhelm features with coordinated zergling swarms.

Improve insights to make smarter decisions by tapping into real-time data with retrieval-augmented generation (RAG).

Interactive LLM Chatbot that constructs direct and transitive software dependencies as a knowledge graph and answers user's questions leveraging RAG and critic-agent approach

Dataflow-guided retrieval augmentation for repository-level code completion, ACL 2024 (main)

The AI and Human powered ticket system to manage your AI based code generation with tickets

Unofficial, Google‑free reimplementation of MLE‑STAR: a lightweight, local‑friendly multi‑agent ML engineering pipeline that uses OpenAI‑compatible LLMs (OpenRouter/Ollama) and DuckDuckGo search to generate, debug, refine, ensemble, and submit Kaggle‑ready solutions—AutoML for tabular tasks with code generation and iterative refinement.

MAS (Multi Agent System) Coordindator to Work with General-purspose and Users Management Agents

A collection of environments and reference agents for planning and reinforcement learning research in partially observable, multi-agent environments.

[TPAMI] Interaction Pattern Disentangling for Multi-Agent Reinforcement Learning

Python implementation of Space-time A*, Prioritized Planning, & Conflict-based Search for Multi-agent Path Finding (MAPF)

Codes of GoMARL accompanying the paper "Automatic Grouping for Efficient Cooperative Multi-Agent Reinforcement Learning"(NeurIPS 2023). GoMARL is a domain-agnostic MARL method that learns automatic grouping for efficient cooperation by promoting intra- and inter-group coordination.

Create configurable Claude Code agents as ComfyUI nodes. Assemble agentic workflows that can leverage all the existing possibilites of ComfyUI.

5th place solution repository for Kaggle CSIRO Image2Biomass, including the experiment workflow used with Claude Code and Codex.

Automated pipeline for generating, verifying, and preparing high-quality function calling datasets for model fine-tuning. Reduces manual annotation through multi-stage validation and conversion to training formats. Built for Azure ML with permissive license models (Phi3.5 MoE, Phi4 etc).

Python client for SuperAGI agent framework

Autonomous AI dev team — 9 Claude agents run as an agile team: plan sprints, write code, review, test, deploy. Runs 24/7 via cron.

A local browser automation agent based on Microsoft Fara-7B model optimized for LM Studio inference.