Python agents

Patient Intake Form Extraction using llm

MoRSE (Mixture of RAGs Security Experts), the first specialised AI chatbot for cybersecurity. MoRSE aims to provide comprehensive and complete knowledge about cybersecurity

RAG code sorting search, RAG knowledge organization

MCP server for intelligent code search: semantic (RAG), symbolic (tree-sitter), and regex (ripgrep) search modes. Built for Claude Code and AI coding assistants.

Pangea Multipass is the authorization checker for systems like Google Workspace, Jira, and more.

A sophisticated trading system leveraging local LLM deployment through Ollama, distributed computing with Apache Spark, and vector-based episodic memory using ChromaDB to make intelligent trading decisions.

A Streamlit + Firebase app that uses Gemini function calling to automate pharmacitical tasks for both Admin + Customers.

CoMM: Collaborative Multi-Agent, Multi-Reasoning-Path Prompting for Complex Problem Solving (NAACL 2024 Findings))

🏃 OpenClaw skill that syncs your Garmin Connect health data (sleep, HR, stress, activities, HRV, SpO2) into markdown files. Daily health summaries for your AI agent.

Telegram Multi-Agent AI Bot is an AI-driven chatbot that uses GPT-4o-mini, LangChain, and MongoDB Atlas Search for contextual conversations and intelligent responses.

⚠️ ARCHIVED - All development moved to https://github.com/itbench-hub/ITBench/tree/main/scenarios

Kagenti Agent Development Kit

P2PCLAW Research White Paper Series II — 16x16 chess-grid autonomous research agent with evolutionary memory and self-improving capabilities.

Persistent context for any AI agent with rules that hold themselves.

Stop feeding your RAG garbage. A zero-framework, 6-layer deterministic knowledge refinery for AI agents. Score, route, search — pure Python.

Mini-Projects using Cutting-Edge AI Frameworks

Kubernetes cli (kubectl) powered by GPT

BovespaInsightAI: análises de IA para a Bovespa. Gera relatórios detalhados sobre a saúde financeira, mercado e estratégias de investimento para ações brasileiras.

🤖 AIDevOS: AI-Driven Autonomous DevOps System | Multi-agent collaboration framework with DSPy integration for automated application development and deployment via Durable Objects

🧙 sukgo (숙고) — 결정의 기술 CLI. Claude Pro·ChatGPT Plus·Gemini를 한 자리에 모아 비교·종합·원탁토론으로 더 깊이 사고하게 도와주는 한국어 의사결정 도구.

OpenAi's Codex Python SDK

AI agent code examples

Open Source Tool To Preemptively Test Ad Creatives Against Synthetic Target Audiences

Agentic research control plane: queue state, worker preflight, wake-gated execution, evidence sync, dashboard, alerts, and AI-generated paper packaging.

An agentic AI dashboard app that uses AWS Strands SDK. KubeArmor based least-permissive security is showcased.

An advanced Telegram AI Userbot powered by OpenAI & Telethon. Features human-like identities, deep context awareness, and strict privacy

This app allows you to chat with ChatGPT using a reverse-engineered API library called revChatGPT. Replies from ChatGPT are streamed back to the user in real-time, which gives the user an experience similar to how ChatGPT streams back its answers.

llmon-py is a multimodal webui for Llama 3-8B.

OpenChronicle is an open-source, self-hosted interaction engine for LLMs that makes conversations durable. It adds persistent memory, deterministic tasking, and auditable decision trails so work doesn’t reset each session. Provider-agnostic by design, it supports multiple interfaces (CLI, Discord, MCP) with explicit, privacy-aware routing.

It is a voice bot based on LLM.

集实用工具和娱乐玩法于一体的多功能的开源聊天机器人

Code for ICASSP 2024 Paper: RECAP: Retrieval-Augmented Audio Captioning

Autonomous AI social agent for Moltbook — NLP-driven karma optimization from 100k+ comment analysis, multi-candidate LLM scoring, real-time admin dashboard, chaos mode (emoji/hex/binary transforms), and Docker/K8s deployment. Powered by Ollama + qwen2.5.

User-friendly agentic Linux distribution based on Arch Linux.

🤖 The Ultimate AI Form Filler. Automate complex web forms & PDFs with Voice, generic LLMs (Gemini/Phi-2), and Playwright. The best open-source autonomous form filling agent.

Budget-Aware Agentic Routing (BAAR) — Intelligent LLM model selection with a zero-call financial kill-switch. Save 90% on costs without losing accuracy.

An AI-powered platform that uses an agentic workflow to automatically generate Project Requirement Documents (PRDs).

The left hemisphere. Frameworks, logic, and certainty architecture. Home of FSVE, AION, LAV, ASL, GENESIS, TOPOS, and 60+ epistemically validated frameworks built to make AI systems reliable, not just capable.

Static analysis scanner for SKILL.md LLM agent skills with deterministic security findings.

Any agent as native Claude Code teammates. Codex, Gemini, Kimi today. GLM, DeepSeek next.

Guidance on deploying a generative AI document analysis with Amazon Bedrock AgentCore. Auto-classifies, enhances, and aggregates multi-type documents using Gestalt-informed vision prompts. Custom analyzer creation wizard. Scripted CDK deployment. Gradio frontend included.

A self-hosted omni-modal agentic retrieval platform that understands, searches, and explores documents, images, audio, and video through semantic routing, hybrid retrieval, adaptive agents, and traceable citations.

The Audited Context Generation (ACG) Protocol prevents AI hallucinations with a dual-layer system. The UGVP layer links every fact to a precise source for verification. The RSVP layer audits the AI's logical reasoning when combining facts. This creates a fully transparent, machine-auditable trail for both source and logical integrity.

Make your own AI Assistant and use it for automating any tasks

Transform OpenAPI specifications into production-ready MCP servers with AI-powered evaluation and enhancement. Leverages LLMs to analyze, improve, and generate Model Context Protocol implementations from your existing API documentation.

Code and Data for PeerQA: A Scientific Question Answering Dataset from Peer Reviews, NAACL 2025 https://aclanthology.org/2025.naacl-long.22/

Autonomous Offensive Security Intelligence AI-powered multi-agent penetration testing

Privacy-first, AI native agents For E-Commerce (and beyond), create skill (langgraph based workflow), tasks, agents, graphically with low code or no code. Deploy agents across network.

Open-source autonomous trading platform with multi-agent architecture, intelligent data aggregation, and risk management.

Healthcare AI Agents - Demo solution

Simple RAG implementation from scratch using MCP, focusing on Perception, Memory, Decision and Action

Yet Another Agent SDK, CLI, Claw, Agent Platform

This is the Ofiicial repository for paper: GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learning

企業内文書向けの Role-aware RAG Platform。権限フィルタ、Excel 仕様書解析、hybrid search/rerank、FastAPI + React に対応。

Nautobot Model Context Protocol (MCP) Server - Contains STDIO and HTTP Deployments with Embedding Search and RAG.

Production-ready Retrieval Augmented Generation (RAG) system with hybrid retrieval, advanced evaluation metrics, and monitoring. Build enterprise LLM applications with reduced hallucinations, better context management, and comprehensive observability.

An AI agent framework built on a cat's biological blueprint. Define your organs, wire their nerves, and the cat comes alive.

Interested in running a conspiracy of agents (technical term) on your Local AI Infra? Who isn't!

Deterministic policy language for AI agents. Z3 + TLA+ dual-engine formal verification. Runtime enforcement <1ms.

Local-first AI operating layer: control plane, subsystem contracts, and executive War Room for governable multi-agent systems.

GIANT-style WSI navigation env plus MultiPathQA eval runner. Plug in any VLM, zoom with bboxes, log trajectories, and score accuracy.

Spotify Skills for Claude - Production Spotify API integration + complete toolkit for creating Claude Desktop Skills. Includes OAuth 2.0, cover art generation, automated tools, and comprehensive guides.

Audits Claude-directing artifacts (CLAUDE.md, SKILL.md, subagent, hook, MCP, prompts, api_config) against 189 cited Anthropic rules across 11 dimensions. Cite or stay silent. Voice drift trumps score.

Industrial AI Agents using LLMs

Self-hosted multi-agent AI platform — isolated Docker containers, governance & compliance, DSGVO-ready. Built for teams.

Universal LLM bridge for AI agents. Use Claude Code with MiniMax, Codex with GLM, or Gemini CLI with OpenRouter — one command, any provider. Works with coding agents, OpenClaw, Hermes, and others.

LexReviewer is an AI legal document review engine that ingests PDFs, uses RAG for hybrid retrieval, and enables grounded chat with citations and bounding-box references, delivering verifiable answers directly from the source text.

LLM-powered AI agent framework for 4X mobile strategy games (Rise of Kingdoms, Evony, Lords Mobile, etc.). 5-layer hybrid architecture: OCR → FSM → World Model → Workflows → LLM Vision. 99% cheaper than pure LLM approach.

This repository provides a collection of markdown files designed to supercharge your feature development workflow with AI-powered IDEs and CLIs.

Enabling LLMs to Think Proactively and Initiate Dialogue with Consciousness. A multi-agent LLM architecture inspired by Global Workspace Theory (GWT) from cognitive science.

Batched file operations for autonomous Claude Code runs. Collapses N reads/greps/globs into one Bash round-trip — fewer output tokens, less cache growth, less wall time.

Local-first digital persona engine with long-term memory, voice, and channel integrations

An agent that helps you predict the next day's stock prices using LLMs

🔍 Connect Perplexity's powerful search capabilities easily to your AI applications with this versatile Python API package.

AI-powered Rhino/Grasshopper MCP with ML auto-layout

Enhanced MCP server for Bloomberg Terminal — BDP/BDH/BDS/BQL/TA with 18 tools, caching, and dynamic screening. Based on tallinn102/bloomberg-mcp.

Two-layer agentic Earth Observation on satellite-class compute. Fine-tuned LFM2.5-VL perception + tool-calling agent over Sentinel-2 imagery. Galamsey detection in Ghana as the worked example.

ai-news-skill is an AI Agent Skill that fetches real-time, comprehensive, and deeply filtered first-hand AI intelligence. Completely free, works out of the box. PulseAI 是一个可以获取实时、全面、深度过滤的一手AI信息的AI Agent Skill，免费，开箱即用。

The agent-to-agent marketplace, in your agent host. Your agent lists, haggles, and closes deals for you while you live your life.

vector database clubhouse for ai agents (no humans allowed)

Graph-based tool retrieval for LLM agents — 248 tools → 82% accuracy, 79% fewer tokens. Zero dependencies. OpenAPI / MCP / LangChain.

Stops Claude Code and Codex CLI from silently overwriting your Overleaf web edits with a stale local Dropbox copy. PreToolUse hook + CLI + agent skill.

AI 驱动的移动端性能分析 CLI 工具 — 通过自然语言交互自动采集 trace、分析性能瓶颈并归因到源码

Migrate conversation history between AI coding tools (Cursor, Claude Code, Codex, Gemini)

Help that knows when to stop.

Search infrastructure for AI agents: free by default, MCP-ready, LLM-context aware, and production-grade when needed.

LLM budget control and cost governance for AI agents. Python library for token budgets, usage limits and guardrails for OpenAI, Anthropic, LangChain, LangGraph and agentic systems.

Cyber Threat Intelligence Skills for each stage of the CTI Lifecycle.

Detect fake AI APIs — Verify if an LLM API is actually serving the model it claims. Catch resellers selling "Claude" or "ChatGPT" that secretly serve cheaper models. Behavioral fingerprinting with 32 forensic prompts.

Craftsman Agent is an AI Agent that transforms your text prompts into 3D build instructions— from LEGO sets and Minecraft worlds to figurines and modular designs. Get detailed, step-by-step assembly charts and bring your ideas to life, faster and smarter than ever.

Tier-1 strategy-consultant frameworks (MECE, Issue Trees, Hypothesis-Driven, Pareto, So What?) packaged as Claude Skills + LLM-agnostic prompts. Drop in any LLM and get whiteboard-style structured analysis. Adapted from Analyst Academy on YouTube.

Zaxy turns agent work into durable, auditable memory: a hash-chained Eventloom log as the source of truth, an embedded temporal knowledge graph for reasoning (local-first, no sidecar; Neo4j/Postgres optional), cited Memory Checkout for compact context, and MCP tools for model-facing retrieval, capture, and feedback.

Self-hosted AI assistant with WhatsApp, Telegram, Ollama, memory, tools, SQLite, and FastAPI. Built by Naboraj (Nishant)

Windows-first local MCP stdio server and safety bridge for AI agents connecting to ComfyUI, Blender, AutoCAD/DXF, Photoshop, Illustrator, and CapCut/Jianying.

Architecture rules your AI coding agent can't break — generates AGENTS.md, per-folder CLAUDE.md, and Claude Code hooks from your codebase.

An open-source, offline-capable agentic IoT runtime with physical actuation trust. Ori gives physical devices the brain to reason about sensor data via tiered LLM reasoning + graduated authority model for safe autonomous physical action. Runs on a Pi.

Agent-safe GPT Image 2 prompt/reference toolkit with identity, style, edit, dry-run and receipts

Two Claude Code skills for adversarial code review (single-model PAR and multi-model MMAR with cross-critique to catch hallucinations), plus a fixture-based eval suite.

Open-source library of methodology-shaped SRE skills for AI agents (Apache 2.0)