Python agents

ASTRA: Automatic Sizing of Transistors with Reasoning Agents

🌍 Leaderboard Repository for "AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agent", ACL2024

An autonomous multi-agent storytelling framework for Minecraft. Features an LLM-driven World Narrator and intelligent agent swarms capable of PvP, building, and survival using a hybrid Python (Reasoning) and Node.js (Execution) architecture.

An advanced multi-agent system for clinical decision support, enabling deep phenotypic prediction by integrating hierarchical multi-modal deviation maps and non-tabular electronic health records.

A benchmark for agentic tool creation.

Calliope is an experimental agentic framework that brings modern AI tools like generative AI (large language models and image generation models), computer vision, and vector databases to bear in the creation of interactive art works that dynamically produce images, video, text, and sound.

A fully free, open-source 4-agent AI council built with LangChain + Ollama. Natural group-chat conversations, zero API costs, runs entirely on your machine.

This project brings together the seamless interactivity of Streamlit and the advanced language capabilities of OpenAI's GPT-3 to create a user-friendly and intelligent chatbot. Users can ask questions, seek assistance, or simply engage in a friendly conversation, and the chatbot responds with contextually relevant and human-like answers.

AI-powered tool that transforms complex research papers into clear, customized explanations. Uses LLMs (GPT-4, Claude, Gemini) to provide adaptive insights from high school to expert levels. Built with Python, Streamlit, and LangChain. Ideal for students, researchers, and lifelong learners. 🌟 Star us for updates! 🤝 Contributions welcome!

🤖 Agentic RAG - An intelligent document assistant with semantic chunking, hybrid search (vector + BM25), multi-format support (PDF, Word, CSV, Excel), and multi-channel bots (Telegram, WhatsApp). Built with FastAPI, React, PostgreSQL/pgvector, and LangChain.

A Retrieval-Augmented Generation (RAG) based question-answering proof-of-concept (PoC) that enables users to query documents using natural language. This system leverages local LLMs via Ollama for enhanced privacy and performance and features a chat-based interface. Built entirely with Python, including both backend an frontend.

Automated AI video generation factory. Create faceless YouTube content from prompts with LLM scripts, multi-character TTS, and automated publishing. Built with FastAPI, LangGraph, and Next.js

SQL and Python Agent for your MySQL databases.

Turn a single prompt into a complete system architecture. ArchAgent is a multi-agent AI pipeline that takes your project idea and outputs a production-ready architecture document, tech stack decisions with reasoning, cost estimations, and a visual system diagram — all in seconds.

LangChain Agent × Ollama 本地部署大模型

MCP Tools Usage From LangChain ReAct Agent / Example in Python

Open-source AI agent orchestration platform. Multi-agent discussions, visual workflows, 30+ tools, human-in-the-loop governance.

LLQuery : Your Conversational SQL Bridge Harness the power of natural language to effortlessly craft SQL queries with LLQuery.

Langchain integration for Salesforce CRM, providing tools for SOQL queries, object schema inspection, and CRUD operations through Langchain's framework

A research framework for implementing and evaluating poisoning attacks on Retrieval-Augmented Generation (RAG) systems, enabling the study of their security vulnerabilities.

Multi-agent marketing automation system built with CrewAI --> generates market research, strategy, content calendars, social posts, blogs, and SEO-optimized content using structured Pydantic outputs.

TgramBuddy: A Python Telegram bot solution leveraging aiogram, OOP, and efficient dependency management with uv, containerized by Docker.

多UAV-USV协同巡检系统 - 强化学习调度 + HCA-A* + 2-opt优化

Multi-agent local SEO landing page generator powered by Claude. 5 agents collaborate to produce production-quality city landing pages with JSON-LD schema. Driven by Claude Code in VS Code.

IntentusNet - Deterministic execution infrastructure for agent and distributed systems, enabling reproducible workflows, reliable intent routing, transport abstraction, and transparent operational control.

Self-hosted agent platform with inference routing across local and cloud hardware, multi-model benchmarking, NVIDIA Openshell sandboxing and STAMP'ed policy-governed agent runs.

Run multiple Agent Zero agents in parallel. Fan-out tasks, token budgets, shared memory, model routing.

[IJCAI 2025] Constructive Conflict-Driven Multi-Agent Reinforcement Learning for Strategic Diversity

Agentic AI–driven stock analytics leveraging Phidata, Google Gemini 2.0 Flash, and Yahoo Finance. Features real-time data, interactive Plotly charts, and a Streamlit dashboard for comprehensive, actionable market insights.

SARAN: Shallow Auto-Regressive Attention Network

AI-powered credit assessment for Indian MSMEs using OCR, WhatsApp analysis, and 7 specialized agents. Processes informal documents into bank-grade credit reports in 90 seconds.

SEO MCP server for kwrds.ai

Voice chatbot with voice+screen output to show that "not everything needs to be spoken"

Official repository for the paper 'Meta-Reasoning Improves Tool Use in Large Language Models'.

Live Memory MCP Server — Shared working memory for collaborative AI agents (Memory Bank as a Service)

A rust optillm port, for use with just-ever/code (or at least my fork of it)

Jarvis AI Voice & Text Assistant – A Python-based desktop AI assistant with voice commands, text chat, translation, news, music, image generation & modern Tkinter UI.

Open-source AI VTuber for Twitch streaming with smart home integration, TTS/STT, and autonomous capabilities. The transparent alternative to closed-source AI assistants.

Your personal AI assistant that runs entirely offline — with voice control, memory, web search and full system interaction. No cloud. No tracking. Just you and your AI.

This Telegram bot was created as a fun and interactive chat companion for my daughter, Argi. It uses the Anthropic API with the Claude 3.5 Sonnet model to generate responses and can switch between various scenarios, providing a diverse and engaging conversational experience.

An autonomous blog generation system powered by Claude AI. Processes topics from a queue, outputs structured JSON content blocks to Supabase, with optional Shopify sync and AI image generation.

Autonomous AI agent for incident response — triage, RCA, remediation, post-mortem, with human approval

AI gaming companion — screen capture from a separate device, Claude vision analysis

AI-powered codebase analysis platform using static analysis, graph traversal, vector retrieval, and LLM review agents.

A production ready AI agent framework built on finite-state machines (FSM).

End-to-end personal knowledge RAG pipeline with OpenClaw + Lark (Feishu) agent integration

An intelligent health monitoring and recommendation system that analyzes wearable device data using Retrieval-Augmented Generation (RAG), simulated vitals, and locally deployed Large Language Models (LLMs). Provides personalized, real-time, evidence-based insights through time-series trend analysis and medical context awareness.

CLI that helps with docs splitting, embedding and exposing them in a seamless manner

对任意it项目生成生成 ai上下文到一个单独文档，传给ai大模型或者知识库，把生成的文档传给 google ai studuio ，对大型编程项目的代码生成准确率，效果远远地吊打在 cursor trae qoder claude-code gemini-code-assit 等ai ide写代码

Historical War Simulation Based on LLM and KG.

Hybrid AI system that uses OCR, ML, and RAG-powered LLMs to turn legal documents into plain-English explanations with risk flags.

[CIKM 2024] Incorporating Appraisal Theory and Emotion Support Strategies for Empathetic Response Generation

The open-source benchmark for LLM memory decay. Measure how Naive, RAG, Chunked RAG, Cascading, and SummaryMemory degrade over 100 conversation turns. Ebbinghaus forgetting curves, 5-provider LLM eval, multi-seed CI. No API key needed.

🧬 RAGIX: Local-first development assistant making LLMs behave like disciplined engineers – Unix-RAG retrieval, sandboxed execution, MCP-compatible, fully auditable

AI-powered multi-agent system for automated proposal generation using context-aware retrieval and persona-based writing.

🚀 Production-ready modular RAG monorepo: Local LLM inference (vLLM) • Hybrid retrieval with Qdrant • Semantic caching • Docling document parsing • Cross-encoder reranking • DeepEval evaluation • Full observability with Langfuse • Open WebUI chat interface • OpenAI-compatible API • Fully Dockerized

Automatically discover the best RAGFlow chunking parameters for each document type. Two-phase optimization with LLM-based evaluation.

A comprehensive cookbook featuring 10 practical AI agent examples using Strands Agents SDK and Amazon Bedrock

dynaprompt: handle your prompts better

TreeRAG: Hierarchical RAG platform that indexes dense PDFs into logical trees for precise retrieval, 100% traceability, and LLM-guided deep traversal. Production-ready with enhanced security and streaming indexing.

High-recall conversational memory retrieval. 93.9% LoCoMo, 98.4% R@5 on LongMemEval — no LLM required. Local-first, cloud-ready.

MCP server for Autodesk Flame — clip control, timelines, and effects with RAG search, anti-hallucination safety, and self-learning

Intelligent AIOps platform with multi-agent architecture, knowledge graph, and time series prediction

Dive into the brand new OpenAI Assistant API, focusing on tools like retrieval and functions.

Self-evolve Gemini CLI instructions, commands, and skills via the gemini CLI itself — GA + GEPA/DSPy, with hard gates before apply.

CLI tool to format codebases for DeepSeek R1 and Claude 3.5

PaperDoctor 是一个用于科研论文审阅与修改的 AI Agent。

BACH v3.12.4-earth: local-first text-based OS for LLM agents with SQLite memory, 113+ handlers, 550+ tools, MCP servers, scheduler and multi-agent orchestration

Welcome to AI Resources! 🎉 This repository is your ultimate toolkit of evolving instruction sets for a cutting-edge AI system designed to revolutionize code and content transformation. Perfect for developers, writers, and creators.

Lightweight self-hosted AI agent platform. Simple / Pipeline / Handler execution modes, hybrid RAG, knowledge graph.

Persistent memory for Claude Code — invisible self-annotating metadata, semantic search, cross-project retrieval

Persistent memory for Claude Code & Codex CLI. Capture knowledge from sessions, compile it into a wiki, and inject the right context back into future work.

Bayesian memory that learns from feedback for LLM agents

🔬 Parallel AI research orchestration tool - Deploy multiple cursor-agent workers simultaneously to conduct comprehensive research with specialized focus areas and intelligent synthesis.

Local-first persistent memory for AI agents — store, recall, and consolidate knowledge across sessions using FAISS, SQLite, and any LLM. MCP server for coding agents and any MCP client.

Multi-agent AI trading framework cho XAUUSD scalp M15 - kết hợp Kronos ML forecast + TradingView signals + DeepSeek council 8 agents.

a collection of examples and reference implementations for working with the Timbr LangChain SDK, LangGraph, and GraphRAG. The examples demonstrate how to use Timbr's ontology-driven semantic layer, natural language interfaces, and graph-based orchestration to build advanced data and knowledge applications

OmniClaw by Long Leo (Ha Dinh Long) — a Claude Code & Antigravity agent OS governed by 8 immortal Python daemons: zero-trust Git guard, 3-tier MemPalace memory, no agent free-will. Not affiliated with other projects sharing this name.

Official implementation of the paper: "GraphDx: A Cost-Aware Knowledge-Enhanced Multi-Agent Framework for Sequential Diagnosis".

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

Local web GUI for Claude Code and Codex with builder/breaker loops, manual cross-agent review, and Git diff timelines.

1st Place Winner (General Judge) - Datadog Self-Improving Agents Hack. Two identical AI agents play Split or Steal. No pre-programmed betrayal. They discover deception on their own. Built with @evancorrea.

Autoresearch Workflow for Designing Chemical Composition Descriptors

An AI-powered Multi-Agent Research Assistant built with Generative AI, Agentic AI, LangGraph, RAG, Persistent Memory, FAISS, and FastAPI. It supports multi-format uploads, intelligent PDF analysis, and expert-like Q&A via Researcher, Summarizer, Critic, and Editor agents—offering deep, contextual, and interactive research insights.

Super-Gemini is a local-first, agentic AI system designed to run on Termux combines powerful terminal emulation with an extensive Linux package collection. Android and act as a universal developer + productivity agent.

An Intelligent Biomedical Semantic-Retrieval System

RAG-in-a-Box: Zero-Configuration Self-Building Agentic RAG System

66 curated Claude Code skills for Flutter / Dart development, testing, release, and AI-assisted design-to-code pipelines.

Open legal intelligence for Mexico. 30,000+ laws across federal, state, and municipal sources · 3.5M+ searchable articles · trilingual UI (Spanish · English · Classical Nahuatl) · first-party RAG chat · OpenAPI + SDKs · built so citizens, accountants, lawyers, and AI agents can navigate Mexican law as code.

AI-powered fitness coach with multi-provider support (Claude, GPT, Ollama). PWA with 27 MCP tools, workout planning, and progress tracking.

Zero-config LLM call inspector. See every API call your AI app makes.

An open foundational toolkit providing essential components for building AI agents with minimal dependencies for local (non-HTTP/API) actions

n8n workflow that pipes Wazuh SIEM alerts through Claude Haiku for AI triage. ~$0.001 per alert. Slack output with risk assessment + investigation commands.

Dossier is a structured job search system built on top of Claude Cowork. The goal is simple: reduce friction, increase consistency, and make decisions based on your actual profile, not generic AI output.

Rubric-driven AI homework grading system built as a Claude Code Skill. Score student submissions with CoT reasoning, bias mitigation, and PDCA quality cycle.

Test lab for RAGdrag — vulnerable RAG target, sample results, and test suite

Retrieval Augmented Generation containerized for Nvidia DGX Spark

GraphMinds: A Python project exploring graph-based algorithms and AI models.

融合大模型推理与RAG检索增强的诈骗短信甄别系统