GitHub agents

Code for the AI Agents blog series—covering agent design, architectures, multi-agent systems, and evaluation.

Automation Framework using LLM-as-a-judge to evaluate of Agentic AI, RAG, Text2SQL at scale; that is a good proxy for human judgement.

Go from company names / urls to full competitive analysis in minutes using a GPT-4 powered LLM Agent

An extensible RL framework for training LLM agents with advanced search capabilities, built on VERL and supporting state-of-the-art search strategies.

An Android GUI Agent powered by Volcano Engine's vision LLM - automate phone operations with natural language commands

Autonomous Polymarket trading agent powered by Claude Code. Expiry convergence strategy, 94.7% win rate over 2 weeks live testing.

(T-ITS) TrajGen: Generating Realistic and Diverse Trajectories With Reactive and Feasible Agent Behaviors for Autonomous Driving

Multi-agent communication bus

This Guidance demonstrates how to effectively orchestrate multiple specialized AI agents to solve complex customer support challenges through different coordination mechanisms on AWS. Modern customer service environments demand sophisticated handling of multi-step interactions, personalized responses, and seamless access to various data sources.

A collection of objects and behaviors for creating dynamic multi agent interactions. 2D|3D Multi Object Behavior library focused on hybrid system interactions with custom Visualization, Data, and performance features.

Working code repository for the paper "SQL-of-Thought: Multi-agentic Text-to-SQL with Guided Error Correction"

shell script to automation make sub agents of claude.

Bitcoin wallet for AI agents. Stablecoins in, Bitcoin out. Keys in hardware enclaves. Works with OpenClaw, Claude Code, or any agent harness.

TARS is a centralized hub for discovering, creating, editing, and managing Claude Code resources. It provides a visual interface for managing skills, agents, commands, hooks, MCP servers, and plugins across multiple projects with safe apply/rollback operations and profile-based configuration sharing.

This application is built in four stages using infrastructure as code with CDK with Python to deploy. In the first stage, an Amazon Aurora PostgreSQL vector database is set up. In the second stage, the Knowledge Base for Amazon Bedrock is created using the established database. The third stage involves creating an Amazon

Demonstrate end to end Agentic SDLC on Microsoft stack, From a Teams meeting (WorkIQ) intent to automated implementation and then Azure deployment. Powered by GitHub Copilot SDK.

Threat hunting command system for agentic IDEs

Multi-Agent Continuity Layer — Keep your AI agents on a single, observable, handoff-ready work thread

Run AI agents safely on your local machine

Glupe 🙏🌹. Constrain where AI can generate code. Keep the rest of your codebase untouched.

HyperAgency (h9y.ai) is a Sovereign Agentic AI Operating System Network built on federated and distributed architecture, enabling autonomous agents that remember, collaborate, self-improve, and operate across users, organizations, and decentralized ecosystems.

An agent with human in the loop that can search the web for information while bypassing bot detection for private sites.

NTK — 基于信息密度路由的多智能体框架。用最少的 token 做最多的事。

🤖 AI驱动的虚幻引擎智能助手 | AI-driven Unreal Engine Assistant - 通过自然语言控制UE编辑器 | Natural language control for UE Editor with 30+ intelligent tools

Git-commit your AI conversations before they disappear.

⚡ Chat MCP Client for Remote hosted MCP Servers (with Composio) and locally hosted MCP servers. 📡

Turn GitHub's auto-generated release notes into human-readable sentences.

Self-hosted, OpenAI-compatible inference for the agentic era: reasoning LLMs, universal tool calling, and the Responses API alongside embeddings, speech, and image models — many models sharing your GPUs, one gateway. Powered by Ray Serve.

Lightweight HTTP proxy enabling Claude Code to work with OpenRouter (200+ models), OpenAI Direct (GPT-5), and Ollama (local inference). Full tool support, streaming, thinking blocks.

Next.js RAG with PGVector

Lightweight AI agent runtime for homelabs, built around deterministic skills and local LLMs.

The aim of this repository is to simulate a forklift model and to train a Deep Reinforcement Learning agent to control the forklift to achieve various tasks such as autonomous navigation to a pallet placed nearby.

Simple verification experiments codes for multi-agent RL using OpenAI MPE environment

Native iOS app for talking to your OpenClaw agents by voice or text. On-device speech recognition, streaming responses, multi-agent channels.

Official Implementation of RSS2025 Paper: Songyuan Zhang*, Oswin So*, Mitchell Black, Zachary Serlin, Chuchu Fan: "Solving Multi-Agent Safe Optimal Control with Distributed Epigraph Form MARL".

A comprehensive AI-powered storyboard generation system built with Mastra that creates visual storyboards with consistent character appearances across all scenes. Features multi-agent architecture, character consistency, memory management, and professional exports.

credential isolation for AI agents. Agents never see real API keys - structural guarantee, not policy.

Detect hallucinations in LLM responses. Verify every claim against source documents using hybrid STS + NLI. Works with LangChain, LlamaIndex, or any RAG pipeline. pip install longtracer

Your AI is smart, but does it know you? The Open Protocol for Context-as-Code. Don't build just Agents. Build Identities.

NotionGPT is an AI that writes and organizes content in Notion for you.

mcp-rag-server is a Model Context Protocol (MCP) server that enables Retrieval Augmented Generation (RAG) capabilities. It empowers Large Language Models (LLMs) to answer questions based on your document content by indexing and retrieving relevant information efficiently.

NexDR (Nex Deep Research), a leading deep research agent that autonomously investigates complex topics and generates rich, structured reports.

一个基于 LangGraph 和 LangSmith 构建的多智能体AI系统

Argus Automation — SOTA desktop automation for AI agents. Works with Claude Code, Codex, and OpenClaw.

Three working examples showing how AI agent development evolved - from traditional RAG to batteries-included SDKs and skill-based frameworks

An Agentic AI Framework for Network Security & Monitoring

嘉立创EDA & EasyEDA 多模态大模型 AI 助手

Windows-native terminal proxy that eliminates scroll-jumping and flicker for AI CLI tools. Sits between your terminal and Claude Code, Gemini CLI, or GitHub Copilot. Rust, ConPTY, zero telemetry.

A curated guide for LLM-agent-driven scientific research automation — from getting started to the frontier.

MCP server for semantic search using local Qdrant vector database and OpenAI embeddings

Native and Private ML inference engine, embeddings, classification, reranking, search, and text generation. Rust core with C# bindings. No Python, no ONNX, no CUDA.

Unofficial. Splunk MCP server. Implemented in Python and TypeScript/JS. Runs searches, queries Splunk, and outputs data as JSON, CSV, or Markdown for agentic LLM workflows. Includes guardrails for input SPL validation and output sanitization. SSE/stdio transport support. Deployment options: Stdio, local HTTP (SSE), or Docker.

A desktop coding agent with a GUI. Electron + React app powered by Pi Coding Agent. A chat-driven development with sandboxed file editing, git integration, dependency-aware task management, persistent memory, and an iOS/iPad companion for remote control.

基于个人知识库的AI问答系统SSE流/Agent智能体/知识图谱RAG/FunctionCall/历史消息/图片生成/图片理解/Embedding/VectorDatabase/RAG

Give your AI agent a wallet, email, phone number, and instant access to paid MCP tools — web search, image/video/music gen, X search, SMS, voice calls, code execution, and more. Works with Gemini CLI, Claude, and any MCP client.

Scientific paper evidence search for ChatGPT, MCP, REST, Python, and AI agents.

A lightweight, high-performance AI/RAG workspace and autonomous agent framework implemented in Go. Inspired by the Leann RAG backend architecture

A platform for Interactive AI-assisted Hypothesis Generation [ACL 2025]

Markdown-first Notion for AI agents -- pages, databases, blocks, and comments in one call.

PuPu is a lightweight, cross-platform desktop AI client that works with both local and cloud-hosted models. Whether you prefer running models on your own machine or connecting to providers like OpenAI and Anthropic, PuPu gives you a unified, elegant interface — your AI, your rules.

Build use cases with VideoDB

[ACL 2025] iAgent: LLM Agent as a Shield between User and Recommender Systems

Runtime Security for Multi-Agent AI — Website & Documentation

Vigilante is a sandbox-first orchestration layer for coding agents. It isolates every task in a git worktree, enforces strict credential scoping, and gives you full audit logs — so your agents can't burn down production.

An Elixir SDK for Claude Code - provides programmatic access to Claude Code CLI with streaming message processing

对 Karpathy 的 LLM Wiki 理念的一次完整实践测试——让 LLM Agent 自动构建和维护一个结构化、互相链接的 Markdown 知识库。

Official implementation of the paper "Contextual Counterfactual Credit Assignment for Multi-Agent Reinforcement Learning in LLM Collaboration". (by Yanjun Chen)

Detects what your AI doesn't know about your project and fixes it. Post-training gap detection, runtime context injection via MCP.

A Claude Code skill for integrating AI with Jira, Confluence, and Bitbucket. Automate Atlassian workflows using Jira Skills, Confluence Skills and Bitbucket Skills. Supports Cloud & Data Center.

Command-line tool for ClickUp with git integration, GitHub linking, and AI agent support

The open-source AI agent OS. CEO orchestrator, persistent memory, real execution, atomic budgets. Self-hosted. No cloud lock-in.

Jakarta Agentic Artificial Intelligence

High-Performance Encrypted Database for .NET 10 | Embedded + gRPC Server | Vector Search • GraphRAG • Analytics

Revolutionary AI ETL with Medallion Architecture: Zero-touch autonomous & HITL pipelines on Databricks

[RA-L 2026] ORION: Option-Regularized Deep Reinforcement Learning for Cooperative Multi-Agent Online Navigation

The most feature-rich web interface for Pi Coding Agent. Session archive & restore, model picker with search, thinking level control, image paste & attach, slash command palette, full Markdown rendering, mobile PWA, and macOS auto-start. No build step. No framework. 4 dependencies.

Security-first package manager for AI agent skills

Connect AI agents to any API, anywhere. Securely. Open-source gateway that proxies requests, injects credentials automatically, punches through NAT to reach localhost services, and wraps REST APIs as MCP tools. Per-agent isolation. Never expose a raw key.

Agent Identity Protocol - Zero-trust security layer for AI agents. Policy enforcement proxy for MCP with Human-in-the-Loop approval, DLP scanning, and audit logging.

A framework for discovering, compiling, and validating reusable skills for scientific agents.

你的AI虚拟女友“小凡”：一个基于Live2D Cubism SDK和OpenAI API的智能对话助手项目，结合了Live2D虚拟形象、实时WebSocket通信和AI对话功能。

Claude省钱版 — 效果基本一致的情况下可以节约Token消耗约90%，超级省Token，支持OpenAI及国产大模型(DeepSeek/GLM/通义千问)的AI编程助手。本项目基于ClaudeCode源码实现。

Claude Code v2.1.88 源码深度分析笔记 — 架构解析 + 面试八股

AI agent-powered CLI for blog publishing, Naver Cafe, Instagram, X (Twitter) & Reddit automation

MCP server providing semantic Java code analysis for AI agents. Built on Eclipse JDT with tools for navigation, refactoring, search, and metrics.

Get comprehensive insights about any competitor with detailed reports on products, market positioning, and strategy.

An AI-powered agent utilizing LLMs (e.g., Claude) with Function Calling to automate tasks in Blender. Enhances workflow efficiency by integrating intelligent command execution and automation. 🚀

A Go toolkit for building AI agents and applications across multiple providers. Unified LLM client, agent framework with handoffs, tool calling, streaming, structured output, MCP integration, and cost tracking.

Curated list of papers, frameworks, benchmarks, and applications for multimodal AI agents (LLMs, text-to-image, speech, world models, etc.) on mobile and edge devices.

Give your agents a cute office. Multi-device, multi-user, multi-agent collaboration.

Team-API 专为团队、组织、企业打造的多租户大模型 API 网关。具备完整权限管理与审计能力，助力团队更好地使用AI大模型。支持Claude/OpenAI/Gemini等主力模。QQ交流群：1095286563

PRO-V: An Efficient Program Generation Multi-Agent System for Automatic RTL Verification

Robinhood portfolio advisor skill for Hermes Agent, OpenClaw, Claude Code, Codex, and other MCP-compatible AI agents

Claude agent skill: Generate proven ad concepts from 320+ templates. 16 universal angles + category-specific formulas.

Files-first operating system for internet businesses: research, decisions, bets, pushes, and outputs in markdown/git, with an mb CLI and agent workflows that hide the plumbing.

Turn your AI assistant into a tutor. It plans, shows code, reviews, adapts to how you learn and helps you complete your real project on your own

将自然语言描述的 COMSOL 建模需求自动转换为完整 .mph 模型文件的智能 Agent，支持几何、物理场、网格、研究与求解的完整仿真流程。

📰 An automated, multi-agent research ops AI pipeline that gathers daily AI news, synthesizes it into structured briefings, and delivers them to tools like Notion, Obsidian, Slack, and Teams with zero manual effort. It also supports on-demand deep research on any topic using parallel AI agents and a modular plugin ecosystem.

Qt C++ library for working with AI/LLM Providers and MCP