Python agents

Your personal AI CTO: Automates the journey from developer to intern by auditing code, suggesting gap-filling projects, and tracking 'Hirability Scores' through a modular agentic workflow.

Verify claims using AI agents that debate using scraped evidence and local language models.

Spackle lets you expose your MCPs, hooks, tools, and builds to Claude across projects with zero futzing or configuration.

A Knowledge Graph-Infused RAG framework for multi-hop reasoning over Türkiye-related entities. It utilizes Neo4j for graph storage, Wikidata5M for structured data, and Ollama (Qwen 2.5) for agentic spreading activation to outperform traditional text-based RAG baselines.

A flexible, modular framework for researching AI agents that design

A starting point for building custom agentic LLM applications using Open Source tooling and models. Incorporates Ollama, Open WebUI, Langchain, Streamlit, Chroma, & PGVector using Docker and Docker Compose and optionally Codespaces.

AgentArticle is an AI-driven application that streamlines content creation by combining web scraping, NLP, and AI tools to produce high-quality articles for LinkedIn and other social media platforms. It integrates APIs and Selenium automation, making it perfect for professionals, content creators, and marketers.

This repository contains the development of my Final Degree Project titled "Multi-Agent System Based on RAG for Assistance in Diabetes Management." The goal of the project is to design and implement an AI multi-agent system that uses RAG to assist patients with diabetes.

This automated snapshot taker uses Caldera/Blockscout API and GitHub workflow

A research-grounded course on writing high-quality Claude Code skills — 9 modules, 14 universal patterns, deployable assess-skill and write-skill, and a terminal reader. Based on analysis of 8 production repos and 6 peer-reviewed papers.

Claude Code agents move any MongoDB collection into Exasol, discover patterns, and publish a live data app — one prompt, end to end.

A lightweight multi-agent research assistant with a clean UI and visible agent activity

Heureum Agent

Automated newsletter pipeline using agentic AI. Topic research → content generation → formatting. Similar pattern to ScribbleFlow — shows a consistent "local-first AI pipeline" thesis.

Jolo — not exactly YOLO but close enough. Program like it's 2026!

Use the Pearl expert MCP server with the OpenAI Agent SDK

Agent-to-Agent AI Integration: AWS Security Analysis with AI21 Maestro & Strands Agents

An agent that Knows You, Remembers You, and Understands You

A research instrument for studying personality drift in LLM agents. Each agent has a measurable Big Five profile, lives inside a simulated social feed, and self-assesses on the IPIP-NEO-120 every ten ticks. Drop in an API key and a seed file to run your own experiment.

Deterministic firewall for MCP agent tool calls. YAML policies, hash-chained audit, blast-radius scoring, multi-cloud cost intelligence.

This is a 3-days Citi Wealth Global Hackathon #HackToTheFuture project that leveraging AWS services to deploy a streamlit application for automating and streamlining the KYC and SOW verification process..

Integración del prototipo desarrollado por los participantes del laboratorio {DATAR} para la orquestación de agentes autónomos.

This is the fully autonomous ai agent build with OpenAI Agents SDK

Multi-agent, policy-driven AI system for processing sensitive enterprise documents with extraction, analysis, verification, deterministic orchestration, and full audit logging. Designed for regulated environments (banking, finance, insurance).

A real-time, zero-trust governance layer that sits between AI agents and the outside world.

An agent that can use browser to do deep search/research on people while bypassing captchas.

Adaptation of Pokemon Knowledge Graph as LinkML Schema

A multi-agent system built on MCP (Model Context Protocol) architecture providing specialized AI assistants with database analytics, weather information, and meeting scheduling capabilities.

Seamless auth flow for the agentic web

Bizy is the first rules and logic layer that machines, engineers and business stakeholders can directly configure and understand.

System of record for AI tool risk: inventory, policy enforcement, approvals, and audit-ready evidence.

Open protocol and CLI for installing autonomous AI behaviors into any code repository.

End-to-end agentic ACR decision system using RL and GRPO

Multi-agent modular workflow system to perform in-depth research on any prompt.

A proactive AI assistant using Qwen-Agent to learn your writing style and manage Google Calendar & Gmail.

Agentic RAG Chatbot using Python, Pinecone Hybrid Search, Ollama (Llama 3), Redis, and Flask. Features modular architecture and stateless scalability.

A Python framework for N-agent collective learning using Tsetlin Machines and synthetic data sharing, supporting zero-shot sensor onboarding and LLM-driven feature extraction.

The Essential MCP Toolkit for Developers

Multimodal Agentic GenAI Workflow – Seamlessly blends retrieval and generation for intelligent storytelling

An intelligent MySQL AI Agent system that translates natural language into SQL queries and provides insightful responses — powered by Streamlit, LangChain, Gemini, and MySQL.

Agentic RAG asset analysis and recommender

A modular multi-agent AI system that performs deep scientific research using a supervisor-worker architecture. It combines foundational and specialized language models to reason, plan, and execute tasks for document and chart analysis in scientific domains.

A muti-agent Dnd DM that directs you to dive into a fantasy world!

Memory engine for self-evolving AI agent ecosystems. Multi-tenant, multi-space isolation with Bayesian truth maintenance, emotional valence, automatic consolidation, and configurable per-agent personality.

Public version of the Human Agentic Platform

Full Form : Expert Network for Intelligent Guidance and Multi-task Assistance. Using multiple LLMs, for Conversational as well as Functional Request.

AI memory for the physical world — 4D spatiotemporal vector database that remembers where things were, not just what's visible now. Hilbert bucketing · temporal sharding · predict-then-retrieve · novelty detection.

Companion repo for Vibe to Live: Building and Deploying Production Agentic Apps

Demo with FastMCP Bearer Auth and progress indicator

Before there was CLUADE CODE - I wrote this similar tool. Leverage LLM-API to translate high level YAML requirements into function/unit-tested applicaiton

A SQL question-answering tool based on LLM

Research and Course material based on AI Security Research

Enables small function calling models to find and use 484,000 local source code tools from many open source repos

👾 Self-Healing AI Terminal Assistant — 100% local, offline-capable, privacy-first. Multi-tier LLM (llamafile/GGUF), auto-fix with FixNet consensus, 80+ commands. No cloud, no API keys needed.

Your personal AI assistant for smarter social media management — Suzy helps you search, organize, and automate your content across platforms.

Mission Control for multi-agent AI systems. Real-time orchestration, live metrics, and complete visibility. Built with LangGraph, FastAPI, and Next.js.

Brain Protocol — open specification for AI agent workspaces

Web UI for atop rawlog files - parses atop 2.7 to 2.12 across RHEL, Ubuntu, SLES, and Amazon Linux. FastAPI + Chart.js, process drill-down, optional AI briefing and chat.

Self-learning AI standards system - learns from corrections and auto-updates CLAUDE.md

五类记忆库涵盖人机协作中的绝大部分场景，完全本地化实现，定期归档，减少上下文token消耗；模拟人类回忆机制，降低无效读取上下文和输出话术累赘的问题，优化官方的心跳和定时机制；情感交互系统让模型具备链接能力，提供专属且长期的智能伴侣养成体验。未来会经历一段很长的人机协作时期，技巧的提炼终归是阶段性的结果，所以该项目所有机制均开放自定义和开关功能，供当下参考，道的实践比术的技巧更重要。

🤖 Full-stack conversational AI using a Letta (MemGPT) + RAG hybrid architecture for long-term memory, context persistence, and grounded responses. Built with FastAPI, React, FAISS, and MongoDB, featuring Isabella — a personality-driven assistant with document ingestion, structured memory, logging, and a terminal-style streaming chat UI.

AI-Web-Reader-Chatbot-Extension is a smart Chrome extension that lets you interact with any webpage using AI. It scrapes and analyzes the content of the current website and answers your questions in real time. Powered by advanced language models, it delivers accurate and concise answers based only on the page content.

Tars is a voice-activated AI assistant that listens for the wake word "Hey Tars", responds to queries, opens applications, and searches the web.

langchain-moorcheh is a Python package that enables seamless integration between LangChain and Moorcheh.

AI-powered Norwegian grocery shopping assistant using RAG (Pinecone + Llama 3.3). Find prices, store locations, and get shopping advice across Kiwi, Meny, Rema 1000, and more. Deployed on Hugging Face Spaces.

Local AI assistant with a SvelteKit web UI. Multi-phase pipeline (route → plan → generate → validate) on top of llama.cpp, Ollama, or LM Studio. RAG, web search, no API keys, no cloud.

⚡ Interactive AI assistant that transforms YouTube videos into a chat experience using RAG (Retrieval-Augmented Generation). Built with Streamlit, LangChain, and Gemini API.

CLI-based AI coding assistant using AWS Bedrock and Claude models. Features interactive mode, file context, git integration, web search, and image analysis.

Your friendly AI assistant that lives in your terminal.

Твой личный кибер-кентуха. Поясняю за железо, разгоняю скуку и баню душнил. Работаю на Gemini, живу по понятиям.

A powerful, local-first AI orchestration layer that unifies advanced LLM reasoning, real-time voice synthesis, and local system automation. Featuring a Flask-based web interface, this system uses a smart Gemini API rotation strategy with a Groq fallback.

🤖 AI-powered personal finance platform with 📊 real-time insights, 📱 automated expense tracking, and 💡 intelligent financial guidance.

A beginner-friendly offline CLI AI assistant built with Python and Rich. Features smart name memory, basic intent handling, and a modular structure ready for future API integration.

基于 OneBot v11 的可扩展 QQ 机器人框架，内置插件系统、会话管理、调度、指标监控与热重载，支持快速开发与稳定运行。

The fastest, most efficient library for running GGUF models with maximum throughput and zero-config hardware optimization.

Skills based claude code plugins for generating automation on multiple platform.

An AI-powered chatbot aimed to deliver accurate and contextually relevant responses by combining the strengths of pre-trained language models with specialized agents

Context Integrity Verification System - система защиты ИИ-агентов от атак на память и контекст, таких как memory injection, prompt injection, tampering и replay.

AI-powered Pizza Order Bot built with Streamlit and Groq's LLaMA 3. The bot takes pizza orders, handles delivery or pickup, summarizes the order, and responds like a human. Fast, natural, and customizable — ideal for learning chatbot integration with LLMs and real-time user interaction.

Self-hosted Mac assistant powered by Claude Code. Telegram, Discord, Slack, iMessage, Web — same butler, anywhere.

MCP server for code structure analysis across 20+ languages. Tree-sitter powered. Works with Claude Code and Claude Desktop.

ContainMind enables AI assistants to interact intelligently with containerized environments, regardless of the underlying runtime (Docker, containerd, CRI-O, etc.). Through its Model Context Protocol (MCP) interface, AI agents can inspect, monitor, and control containers using natural language

Multi-agent AI code review system that uses consensus between specialized agents to provide high-quality inline feedback.

Local-first macOS coding assistant with a FastAPI backend, Swift workspace app, and floating overlay.

AI assistant with MCP integration, agent skills, knowledge graph, and task scheduling

Educational AI chat client: provider abstraction, token compression & state management in ~600 lines Python. Learn robust AI integration patterns.

Chatbot educativo con IA local (Ollama + RAG) para asistencia académica universitaria. Sistema completo con autenticación, historial de sesiones, dashboard de métricas y diagnósticos de rendimiento..

AI assistant backend for document-based question answering using RAG (LangChain, OpenAI, FastAPI, ChromaDB). Features modular architecture, multi-tool agents, conversational memory, semantic search, PDF/Docx/Markdown processing, and production-ready deployment with Docker.

Python-based voice-controlled desktop assistant with automation, AI responses, system control, and productivity modes.

A GUI tkinter-based application that allows two LLM models to have conversation with each other.

Give Claude Code a memory that persists across sessions and survives context compaction. Zero dependencies, pure Python, 30-second install.

Claude Code security snippets for safer permissions, sandboxing, and shell guardrails.

A Claude Code plugin that turns any project's docs into a persistent, compounding, LLM-maintained knowledge base — with hook-enforced discipline at tool-call time. Inspired by Andrej Karpathy's llm-wiki.md gist.

Prompt-injection defenses for Claude Code. A PreToolUse Bash hook blocks compositional credential-exfiltration shapes (secret read plus network, env dump to network, remote script to shell, reverse shells). A sanitizing MCP server wraps untrusted URLs and files in sentinels, strips invisible unicode, flags jailbreaks.

This is a FastAPI service that provides a RESTful API for the Anthropic Claude AI model using anthropic-sdk-python.

WhatsApp Business API from Claude Code — create templates, send messages, run ops in natural language. Direct Meta Cloud API, no BSP.

CLI benchmark suite for LLM providers and OpenAI-compatible gateways. Measure TTFT, latency, p95, throughput, warmup, and history.