ULIVZ

agents-last-exam. Agents' Last Exam

container. A tool for creating and running Linux containers using lightweight virtual machines on a Mac. It is written in Swift, and optimized for Apple silicon.

cybergym. CyberGym is a large-scale, high-quality cybersecurity evaluation framework designed to rigorously assess the capabilities of AI agents on real-world vulnerability analysis tasks.

browser-harness. Browser Harness | Self-healing harness that enables LLMs to complete any task.

obscura. The headless browser for AI agents and web scraping

otp. Erlang/OTP

verl. verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

trl. Train transformer language models with reinforcement learning.

NYU_CTF_Bench. Python

execnet. distributed Python deployment and communication

pytest-xdist. pytest plugin for distributed testing and loop-on-failures testing modes.

openfang. Open-source Agent Operating System

harbor. Framework for evaluating and improving agents

claude-mem. Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

Nuitka. Nuitka is a Python compiler written in Python. It's fully compatible with Python 2.6, 2.7, 3.4-3.14. You feed it your Python app, it does a lot of clever things, and spits out an executable or extension module.

dbus. D-Bus is a message bus system, a simple way for applications to talk to one another.

supacode. worktree coding agents command center.

knowledge-work-plugins. Open source repository of plugins primarily intended for knowledge workers to use in Claude Cowork

pydantic-ai. AI Agent Framework, the Pydantic way

monty. A minimal, secure Python interpreter written in Rust for use by AI

baoyu-skills. TypeScript

claudes-c-compiler. Claude Opus 4.6 wrote a dependency-free C compiler in Rust, with backends targeting x86 (64- and 32-bit), ARM, and RISC-V, capable of compiling a booting Linux kernel.

pygments. Pygments is a generic syntax highlighter written in Python

ty. An extremely fast Python type checker and language server, written in Rust.

pyrefly. A fast type checker and language server for Python

basedpyright. pyright fork with various type checking improvements, improved vscode support and pylance features built into the language server

cpython. The Python programming language

dspy. DSPy: The framework for programming—not prompting—language models

systemd. The systemd System and Service Manager

sandbox-agent. Run Coding Agents in Sandboxes. Control Them Over HTTP. Supports Claude Code, Codex, OpenCode, and Amp.

beads. Beads - A memory upgrade for your coding agent

gastown. Gas Town - multi-agent workspace manager

SwanLab. ⚡️SwanLab - an open-source, modern-design AI training tracking and visualization tool. Supports Cloud / Self-hosted use. Integrated with PyTorch / Transformers / verl / LLaMA Factory / ms-swift / Ultralytics / MMEngine / Keras etc.

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

httpx. httpx is a fast and multi-purpose HTTP toolkit that allows running multiple probes using the retryablehttp library.

python-markdownify. Convert HTML to Markdown

zerobrew. A 5-20x faster experimental Homebrew alternative

tqdm. :zap: A Fast, Extensible Progress Bar for Python and CLI

Stable-DiffCoder. Stable-DiffCoder is a family of lightweight open-source code DLLMs(diffusion large language models) comprising base and instruct models, developed by ByteDance Seed.

packaging. Core utilities for Python packages

ralph-claude-code. Autonomous AI development loop for Claude Code with intelligent exit detection

agent-lightning. The absolute trainer to light up AI agents.

openwork. The open-source alternative to Claude Cowork (powered by opencode)

Megatron-LM. Ongoing research training transformer models at scale

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

tianshou. An elegant PyTorch deep reinforcement learning library.

mujoco. Multi-Joint dynamics with Contact. A general purpose physics simulator.

atari. AI research environment for the Atari 2600 games 🤖.

textual. The lean application framework for Python. Build sophisticated user interfaces with a simple Python API. Run your apps in the terminal and a web browser.

commit0. Commit0: Library Generation from Scratch

SWE-smith. [NeurIPS 2025 D&B Spotlight] Scaling Data for SWE-agents

ui-ux-pro-max-skill. An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms

oh-my-openagent. omo/lazycodex: The coding agent for tokenmaxxers;the one and only agent harness for complex codebases. For your Codex, for your OpenCode

parquet-tools. A utility to deal with Parquet data

omegaconf. Flexible Python configuration system. The last one you will ever need.

frontier-evals. OpenAI Frontier Evals

verifiers. Our library for RL environments + evals

aiofiles. File support for asyncio

aiohttp. Asynchronous HTTP client/server framework for asyncio and Python

aiodocker. Python Docker API client based on asyncio and aiohttp

docker-alpine. Official Alpine Linux Docker image. Win at minimalism!

alfworld. ALFWorld: Aligning Text and Embodied Environments for Interactive Learning

baml. The programming language for agents

flash-attention. Fast and memory-efficient exact attention

rllm. Democratizing Reinforcement Learning for LLMs

vllm-omni. A framework for efficient model inference with omni-modality models

lanser-cli. [Lanser-CLI] Official Implementation of "Reinforcement Learning from Compiler and Language Server Feedback" (https://arxiv.org/abs/2510.22907)

CodeClash. Benchmarking Goal-Oriented Software Engineering

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

vllm. A high-throughput and memory-efficient inference and serving engine for LLMs

SWE-agent. SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

3FS. A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

minikube. Run Kubernetes locally

AgentBench. A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)

pydantic. Data validation using Python type hints

agentskills. Specification and documentation for Agent Skills

exo. Run frontier AI locally.

a2ui. TypeScript

multi-swe-bench. Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving

UI-TARS-desktop. The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra

click. Python composable command line interface toolkit

pytorch. Tensors and Dynamic neural networks in Python with strong GPU acceleration

iceberg-python. PyIceberg

Trinity-RFT. Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).

spark. Apache Spark - A unified analytics engine for large-scale data processing

verus. Verified Rust for low-level systems code

astropy. Astronomy and astrophysics core library

niri. A scrollable-tiling Wayland compositor.

ttyd. Share your terminal over the web

sandbox-runtime. A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.

arrakis. A fully customizable and self-hosted sandboxing solution for AI agent code execution and computer use. It features out-of-the-box support for backtracking, a simple REST API and Python SDK, automatic port forwarding, and secure MicroVM isolation. Perfect for safely running, testing, and backtracking multi-step agent workflows.

lambda-perf. Lambda Cold Starts analysis, updated daily

dify-sandbox. A lightweight, fast, and secure code execution environment that supports multiple programming languages

valuecell. ValueCell is a community-driven, multi-agent platform for financial applications.

skills. Public repository for Agent Skills

agent-sandbox. agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes.

typebox. JSON Schema Type Builder with Static Type Resolution for TypeScript

swagger-parser. Swagger 2.0 and OpenAPI 3.0 parser/validator

openapi-typescript. Generate TypeScript types from OpenAPI 3 specs

Valdi. Valdi is a cross-platform UI framework that delivers native performance without sacrificing developer velocity.

smolagents. 🤗 smolagents: a barebones library for agents that think in code.

mobilerun. Automate your mobile devices with natural language commands - an LLM agnostic mobile Agent 🤖

amplifier. Shell

mcporter. Call MCPs via TypeScript, masquerading as simple TypeScript API. Or package them as cli.

tinyproxy. tinyproxy - a light-weight HTTP/HTTPS proxy daemon for POSIX operating systems

llvm. Library for interacting with LLVM IR in pure Go.

how_to_fix_your_context. Jupyter Notebook

nanochat. The best ChatGPT that $100 can buy.

exa-mcp-server. Exa MCP for web search and web crawling!

unsloth. Unsloth is a local UI for training and running Gemma 4, Qwen3.6, DeepSeek, Kimi, GLM and other models.

sora-mcp. An MCP server to use Sora video generation APIs

xyflow. React Flow | Svelte Flow - Powerful open source libraries for building node-based UIs with React (https://reactflow.dev) or Svelte (https://svelteflow.dev). Ready out-of-the-box and infinitely customizable.

trae-agent. Trae Agent is an LLM-based agent for general purpose software engineering tasks.

rstest. The JavaScript testing framework powered by Rspack

sandbox. All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.

Open-Interface. Control Any Computer Using LLMs.

rslint. High-performance, ESLint-compatible linter for JavaScript and TypeScript.

m3-agent. Python

mcp-hub. A centralized manager for Model Context Protocol (MCP) servers with dynamic server management and monitoring

langextract. A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

agent-tars-glm-4.5-adapter. 让 GLM-4.5 完美适配 Agent TARS 系统的高性能适配器 - 解决 toolcall 兼容性，提供智能 fallback，极低成本享受顶级 AI Agent 体验

SandboxFusion. Python

WasmEdge. WasmEdge is a lightweight, high-performance, and extensible WebAssembly runtime for cloud native, edge, and decentralized applications. It powers serverless apps, embedded functions, microservices, smart contracts, and IoT devices.

coze-studio. An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.

agent-browser-sdk. Agent browser SDK.

Kimi-K2. Kimi K2 is the large language model series developed by Moonshot AI team

vchart-mcp-server. A Model Context Protocol (MCP) server for the @visactor/vchart that enables AI assistants to generate interactive charts and visualizations.

macos. MacOS inside a Docker container.

Crawler_Illegal_Cases_In_China. Collection of China illegal cases about web crawler 本项目用来整理所有中国大陆爬虫开发者涉诉与违规相关的新闻、资料与法律法规。致力于帮助在中国大陆工作的爬虫行业从业者了解我国相关法律，避免触碰数据合规红线。

MediaCrawler. 小红书笔记 | 评论爬虫、抖音视频 | 评论爬虫、快手视频 | 评论爬虫、B 站视频 ｜ 评论爬虫、微博帖子 ｜ 评论爬虫、百度贴吧帖子 ｜ 百度贴吧评论回复爬虫 | 知乎问答文章｜评论爬虫

OpenHands. 🙌 OpenHands: AI-Driven Development

a-share-mcp-is-just-i-need. Python

Seed1.5-VL. Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning, achieving state-of-the-art performance on 38 out of 60 public benchmarks.

VAGEN. World model reasoning RL for multi-turn VLM agents

scrapybara-python. Scrapybara Python SDK

HyperAgent. AI Browser Automation

ffmpeg-mcp. An MCP server for FFmpeg

deepwiki-mcp. 📖 MCP server for fetch deepwiki.com and get latest knowledge in Cursor and other Code Editors

GSAP. GSAP (GreenSock Animation Platform), a JavaScript animation library for the modern web

Bagel. Open-source unified multimodal model

codex. Lightweight coding agent that runs in your terminal

claude-code. Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

ripgrep. ripgrep recursively searches directories for a regex pattern while respecting your gitignore

cline. Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Roo-Code. Roo Code gives you a whole dev team of AI agents in your code editor.

PocketFlow. Pocket Flow: 100-line LLM framework. Let Agents build Agents!

ysymyth.github.io. SCSS

fastapi_mcp. Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth!

simple-evals. Python

genkit. Open-source framework for building AI-powered apps in JavaScript, Go, and Python, built and used in production by Google

awesome-mcp-servers. A collection of MCP servers.

duck-duck-scrape. 🔎 Search from DuckDuckGo and utilize its spice APIs in Node

cua. Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

llms-txt-hub. 🤖 The largest directory for AI-ready documentation and tools implementing the proposed llms.txt standard

cpp-react. C++

firecrawl. The API to search, scrape, and interact with the web at scale. 🔥

debug-router. C++

primjs. JavaScript Engine Optimized for Lynx

lynx-devtool. Debug Lynx On-the-Fly

lynx-stack. The frontend framework and toolchain of Lynx

.github.

streamlit. Streamlit — A faster way to build and share data apps.

OmniParser. A simple screen parsing tool towards pure vision based GUI agent

open-infra-index. Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

turndown. 🛏 An HTML to Markdown converter written in JavaScript

postgres. Postgres.js - The Fastest full featured PostgreSQL client for Node.js, Deno, Bun and CloudFlare

mcp-youtube. A Model-Context Protocol Server for YouTube

browser. Lightpanda: the headless browser designed for AI and automation

DeepSeek-V3. Python

ten-framework. Open-source framework for conversational voice AI agents

repomix. 📦 Repomix is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Language Models (LLMs) or other AI tools like Claude, ChatGPT, DeepSeek, Perplexity, Gemini, Gemma, Llama, Grok, and more.

omnivore. Omnivore is a complete, open source read-it-later solution for people who like reading.

ladybird. Truly independent web browser

json-repair. 🔧 Repair JSON！Solution for JSON Anomalies from LLMs.

hip. HIP: C++ Heterogeneous-Compute Interface for Portability

ColossalAI. Making large AI models cheaper, faster and more accessible

Open-Sora. Open-Sora: Democratizing Efficient Video Production for All

json. Strongly typed JSON library for Rust

hogchat. Chat with your PostHog data

hertz. Go HTTP framework with high-performance and strong-extensibility for building micro-services.

distance. Distance functions to compare vectors

rquickjs. High level bindings to the quickjs javascript engine

pglite. Embeddable Postgres with real-time, reactive bindings.

trpc. 🧙‍♀️ Move Fast and Break Nothing. End-to-end typesafe APIs made easy.

UFO. UFO³: Weaving the Digital Agent Galaxy

axlearn. An Extensible Deep Learning Library

jax. Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

mlx. MLX: An array framework for Apple silicon

gligen-gui. An intuitive GUI for GLIGEN that uses ComfyUI in the backend

dlight. DX-first UI Rendering Library

infinity. The AI-native database built for LLM applications, providing incredibly fast hybrid search of dense vector, sparse vector, tensor (multi-vector), and full-text.

forge. :electron: A complete tool for building and publishing Electron applications

typespec. Java

serde. Serialization framework for Rust

llm-course. Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

lazy-static.rs. A small macro for defining lazy evaluated static variables in Rust.

thiserror. derive(Error) for struct and enum error types

priompt. Prompt design using JSX.

nodejs-mobile. Full-fledged Node.js on Android and iOS

langchainjs. The agent engineering platform