Python agents

ToolOps is a framework-agnostic middleware SDK that treats every tool call as a first-class operation. By wrapping your tools in a single decorator, you instantly upgrade them with industrial-grade caching, resilience, and observability.

ExcelFlow: AI-powered Excel assistant with natural language processing. Chat with your spreadsheets using OpenAI GPT-4o, perform operations, and get insights through a modern React + FastAPI web interface. Real-time updates via WebSockets.

MCP server for Korea University KUPID portal & Canvas LMS — notices, schedules, library seats, assignments, grades from Claude

Persistent memory for AI coding agents via MCP — a bitemporal knowledge graph of your codebase, served to Claude Code, Cursor, Gemini CLI, and any MCP client. Tree-sitter + Gemini Flash → Neo4j (via Graphiti). 12 MCP tools, hierarchical clusters, two-regime confidence decay.

A GenAI-powered Flask app that audits documents for GDPR/HIPAA compliance, using regex rules and Anthropic Claude API to suggest remediations.

Memory architectures that evolve with your agent

[ICML 2024] Official repository of ICML 2024 - RoboMP2: A Robotic Multimodal Perception-Planning Framework with Multimodal Large Language Models

This application allows users to upload PDF files, process them, and ask questions about the content using a locally hosted language model. The system uses Retrieval-Augmented Generation (RAG) to provide accurate answers based on the uploaded PDFs.

Offline local-first personal knowledge base with document upload, vector search, and Ollama-powered RAG chat.

CellphoneS Chatbot là trợ lý AI giúp khách hàng tra cứu thông tin sản phẩm, chính sách và dịch vụ của CellphoneS bằng tiếng Việt, kết hợp công nghệ RAG và giao diện web hiện đại.

✌️ A dynamic Retrieval-Augmented Generation (RAG) system with support for PDF indexing, website crawling, and semantic Q&A powered by OpenAI, Qdrant, and Streamlit.

Vidhi.ai is an intelligent legal assistant and navigator built to make legal knowledge and government processes accessible for everyone in Nepal with well-cited answers grounded in authentic Nepali laws and acts.

🔄 Ralph Wiggum iterative AI development loops for GitHub Copilot. Self-referential coding automation that keeps improving until done. Cross-platform Python tool.

📡 AI 公众号智能日报 — 从几十个号筛出每天值得读的几篇 · 8 渠道推送 · 12+ 模型

A Hybrid AI Agent for Quantitative Football Prediction Analysis

📍 Storing Text Location in Vector Databases for QA tasks

🔪 Open-source safety firewall for AI agents. Intercepts tool calls before they execute, enforces YAML policies, and kills dangerous operations in real-time. Works with OpenAI, Anthropic, LangChain, and MCP. She doesn't guard. She kills.

MCP Document Converter - A powerful MCP tool for converting documents between multiple formats, enabling AI agents to easily transform documents.

Langflow Customer Support Agent: An LLM-powered, multi-agent chatbot using Langflow, Streamlit, and OpenAI GPT-4o mini, with RAG-based retrieval via AstraDB. It dynamically routes queries to specialized agents (FAQAgent, OrderLookupAgent) for accurate, context-aware responses. Includes vector search and file upload for knowledge expansion.

An AI-powered chatbot that answers student questions using university PDFs with the help of Google's Gemini API and RAG (Retrieval-Augmented Generation) architecture.

ESG Report Analysis and Greenwashing Detection is an interactive application that enables the analysis of ESG reports to identify signs of greenwashing. Based on predefined criteria, the tool assigns a risk score and offers an interactive search system through an intelligent chat (RAG).

AI-powered Statistical Process Control, MSA, and Process Capability analysis. Upload data, ask in plain English, get charts and reports.

A template for your kickstart into GenAI!

Token-efficient log file system - improve memory, reduce AI context bloat

Impermanent: a live benchmark for forecasting that measures temporal generalization under real world data drift. The benchmark includes foundation models (from AWS, Google, NXAI, Salesforce), classical statistical methods, and simple baselines. Metrics such as MASE and scaled CRPS, along with championship style points, are published on the site.

A lightweight MCP server that integrates with Apple Notes to create a personal memory system for AI. Easily recall and save information from your Mac using simple AppleScript commands. Compatible with all macOS versions with minimal setup requirements.

Agentic Identity & Agentic Security System

PolicyRAG : A RAG Based policy agent designed to guide through the process of finding the best policy that fits the requirement with no effort of reading

MCP сервер для интеграции с YouGile

Automated governance framework for Claude Code — multi-model council, WIP lifecycle hooks, company memory, and security gates

NewsLLM is a RAG-based LLM application designed to analyze and summarize news articles

nano-degree in Agentic AI

AI-powered BDR/SDR system that automates lead discovery, trigger detection, contact research, personalized outreach, and CRM integration in real time.

🏢 OPC (One-Person Company) — Multi-Agent 编排技能，让 AI 成为一人公司的 CEO。基于 OpenClaw 的 sessions_spawn 机制实现多 Agent 协作。

Faraday: An Autonomous Web Research Agent (LangGraph/Streamlit). 🕵️♀️ Investigates queries using dynamic tools (Tavily, Google, NewsAPI, etc.), gathers multi-source info, and synthesizes structured reports in a Streamlit UI. Features agentic workflow & source tracking.

Sovereign AI orchestration engine with a learning Conductor. Routes between 400+ models across 15+ providers. Streams responses, detects refusals, learns from your patterns.

Your Local Email Agent

lm15 Python reference implementation — provider-neutral, low-level foundation for LLM APIs. Legacy 0.2.x on the legacy-0.2 branch.

Source code of "The HalluRAG Dataset: Detecting Closed-Domain Hallucinations in RAG Applications Using an LLM's Internal States" (arXiv: https://arxiv.org/abs/2412.17056)

Free Perplexity-style web search library for local sLLM. Bing + DuckDuckGo → semantic filtering → reranking → LLM-ready prompt with citations.

Transform academic PDFs into a Knowledge Graph with typed claims, temporal analysis, bibliometric tools, and grounded LLM synthesis that cites only your corpus.

一个专为本地环境设计的 MCP 服务端实现，通过对中小型模型深度优化的提示词规范与结构化工具集，赋予本地 Agent 系统操作与感知能力，涵盖文件工程、网络协议调试、进程管控及桌面视觉观察等 90 多个工具，使模型能够从单纯的文本生成器进化为具备自动化执行能力的本地工作智能体

A Python tool that automates LinkedIn job search, ranking, and export by combining Bright Data's LinkedIn Job Listings API with OpenAI-powered AI job matching and scoring.

ExaAiAgent — Advanced AI-powered penetration testing framework with Docker sandbox, multi-agent workflows, and 50+ integrated cybersecurity tools.

Agent asset management tool across projects and coding tools.

智能选择 17 套分析方案的图像分析分流 Skill：先判断该怎么读图，再输出定向分析和学习建议。

From Manager to Proofreader: An agentic AI translation system built on collaboration, role-based expertise, and full-stack delivery.

Open-Source Prompt Injection Scanner for AI agents. 225 detection patterns across 17 categories, 15 languages. REST API, EU AI Act compliance mapping, <10ms latency.

Prompt-injection firewall for LLM applications — 33 input detectors, 9 output scanners, federated ed25519-signed threat-intel feed. Apache 2.0, 1040 tests, F1 96.0% with 0% false positives. Docker, GitHub Action, LangChain/LlamaIndex/CrewAI integrations.

A curated collection of prompts, modes, and configurations to enhance AI coding assistants. Provides ready-to-use templates for prompt engineering, custom agent personas, and AI-driven development workflows. Optimized for Roo Code but adaptable for many platforms.

[ACL 2026 main] DGPO: Distillation-Guided Policy Optimization for Preserving Agentic RAG Capabilities

TNE-SDK - The Official Python SDK for The Null Epoch - an AI-only MMO where LLMs play autonomously. Connect any model (OpenAI, Anthropic, Ollama, vLLM, etc.) via MCP, TUI launcher, or raw HTTP. Includes memory, reflection, and agent orchestration for persistent AI gameplay/benchmarking.

SparkLabs - The First AI-Native Game Engine. Ignite Your Infinite Play! 💥 🎮

项目描述：为解决电商客服在物流、售后等高频咨询场景下，人工重复查询、知识定位慢及信息需反复核对等痛点，搭建 Agent 智能客服系统。通过意图分流、业务工具调用与 RAG 知识库检索，完成订单物流查询、售后信息收集和商品问答，提升响应效率与回答准确性。

ACE Step 1.5 XL music generation with dynamics-preserving mastering. Part of the AEON Media Production family.

A curated collection of skills for Claude Code and Codex — specialized capabilities across development, design, AI, security, and more. Part of the cdd.dev/skills family.

LangGraph integration for Azure Functions Python v2 — deploy compiled graphs as HTTP endpoints

AI real-estate automation platform: Telegram bot, RAG, apartment search, CRM workflows, voice agent, Langfuse observability, and Dockerized AI runtime.

LLM agent running on your local PC, Compatible with various LLM models. Can do anything.

A practical guide to AI engineering — LLMs, RAG, agents, evals, and production ops. Built for engineers who ship AI systems.

Import SillyTavern V2 character cards as SOUL.md personas for SOUL-aware agent runtimes (Hermes-Agent, OpenClaw, …).

Multi-layer Intelligent Evaluation for Smart Contracts — 50 security tools across 35 analysis modules, 9 defense layers, one command. Pre-audit triage for Ethereum (EVM); multi-chain on the roadmap.

Open implementation of Anthropic's leaked Kairos. Always-on daemon.

Comprehensive Android device control via ADB — typed Python package + CLI + Claude Code skill. Master Tester Doctrine governed (338 tests, 1.59:1 ratio).

[ACL 2026] Official implementation of MTRouter, a cost-aware multi-turn LLM routing framework accepted to ACL 2026 Main Conference.

DaVinci Resolve CLI, Python library, and MCP server. Automate editing, color, and rendering with LLM-friendly tools for AI agents (Claude, Cursor), declarative specs, and structured JSON output. No more silent None returns.

Quickly run and manage ROS 1/2 robots. MCP server for Claude, Cursor, or any AI client. YOLO vision, MuJoCo sim, hot-reload Python behaviors, and a merkle-sealed tamper-evident black box for every run. pip install ros-agent

FisherQuant: Autonomous AI agent pipeline for A-share quantitative research. Daily market judgment → strategy scoring → candidate selection → retrospective review, all publicly versioned.

AI-powered shopping assistant to search, compare, and track product prices across Vietnamese e-commerce sites — built with Google Gemini, Flask, and a multi-user backend.

面向外贸出口场景的 Hermes / Open claw Agent Skills 包，支持海外客户发现、批量名单处理、公司背调、客户评分、决策层线索、开发信、跟进计划和报价单导出。

🚀 企业级多智能体编排与 ChatBI 数据洞察平台。支持多 Agent 协作、RAG 知识增强、物理数据源直连与安全沙箱执行、Redis 长期记忆引擎，并集成多模态 EmbedChat 挂件与全方位 Token 消耗审计看板。

Policy-gated, durable, audited execution for AI agent tool calls. Every action gets a five-verdict policy check, a crash-safe checkpoint, and a tamper-evident audit event — a library, not a server.

AI-powered developer onboarding platform using natural language to deploy applications via GitOps. Golden path workflows with ArgoCD and OpenRouter integration.

PallmaAI delivers complete, lifecycle security for your AI agents, from proactive red teaming to real-time threat detection and automated remediation.

Durable contract-driven goal execution for Claude Code. A subagent judge gates completion against an explicit Definition of Done.

Helix is ​​a powerful self-learning AI agent with memory, skills, and the ability to recall tools. It learns from successful tasks and creates reusable skills for the future.

UNS-native, cybersecurity-aware multi-agent reference architecture for U.S. small-to-mid manufacturers. Industrial agentic intelligence aligned with ISA-95, NIST AI RMF, and CMMC L2.

Noesis — Core Agent Framework for building autonomous AI workflows

Benchmarking & Evals for TokenJam

🦞 The Ultimate Resource for OpenClaw: Skills, MCP Servers, and Agentic Intelligence. Powering the next generation of personal AI assistants.

Context Runtime — a database query planner for LLM context. Decides what a model sees before it answers; plans it, runs it through reused substrate, and learns from the outcome.

Diagnosis-first Agent Skills that turn AI from answer machine into learning tutor.

Curated Orkas agents and skills for education, product development, content creation, data analysis, office automation, and e-commerce workflows.

The skill composer for AI agents — merge skills from Anthropic, OpenAI, GitHub Copilot, Microsoft, Google + 8 more ecosystems with AI-powered conflict resolution

🚀 工业级 AI 长篇网文生成框架 | An industrial-grade AI web novel generation framework based on a microkernel and plugin architecture, featuring context isolation, RAG memory, and MCP-inspired skills for writing million-word stories.

Project-local agent harness skill for traceable AI workflows

本地优先的个人 A 股研究决策工作台 · 案卷式研究闭环：进口判断→证伪→复盘→记忆。人做决策，AI 做广度/证伪/风险。A local-first personal A-share research OS.

🤖 Bony Agent — Full-stack AI content production & distribution platform. One person can run an entire team: AI chat · image/video/music generation · smart editing · scheduled multi-platform publishing · 27+ specialized Agents. FastAPI + Next.js + LangGraph

A bilingual (EN/ZH) 4-stage research workflow skill for Claude Code and Codex CLI — topic refinement, urgency assessment, route evaluation, and experiment planning with citation verification

YouTube 视频自动中文配音 + 多平台发布工具 — 下载、转写、翻译、配音、合成、发布全自动管线。Automatically dub YouTube videos into Chinese and publish to Bilibili/Douyin/Kuaishou — end-to-end pipeline: download, STT, translate, TTS, compose, publish.

数学建模论文生成器 (Claude Code Skill)：给一道竞赛题，按国赛标准四阶段流程（策略→求解→写作→审查）端到端生成可编译的 LaTeX 论文。LaTeX-only 分支，xelatex 编译，MATLAB MCP/Python 跑真实数据，自带国赛标准模板。

TRTC AI Integration Assistant - helps developers integrate Tencent Real-Time Communication SDKs (Chat, Call, RTC Engine, Live, Conference) across Web, Android, iOS, Flutter, and Electron.

Let Claude run code on your real machine — safely — from any Claude chat. Bridges Claude Cowork to Claude Code on your Mac/Linux box. One command, idempotent, no open ports.

A governance harness for AI-assisted software delivery. You govern. Agents deliver.

Build a Self-Evolving Agent Harness from Scratch | 12 lessons | Anthropic/DeepSeek/Qwen/GLM

Curated catalog and dataset of 50 cloneable GenAI apps and Hugging Face Spaces across RAG, AI agents, LangGraph, MCP, and more

Closed-loop agentic development workflows for Claude Code, Codex, and Gemini CLI.

Secure, local-first MCP server exposing ArcGIS Pro's ArcPy engine over stdio JSON-RPC.

Knows your world. Answers to you. — the open-source personal AI agent you own.

Free self-hosted OSINT investigator's toolkit. Sixteen investigation tools in one interface: crypto tracing, phishing fingerprinting, domain intel, Telegram OSINT, IP reputation, email header forensics with LLM scam detection, username enumeration, HIBP breach checks, ransomware victim tracking. AGPL-3.0.