AI Agents

Rust-built autonomous coding agent and self-maintained museum exhibit created with Gajae-Code and LazyCodex, capable of understanding codebases and performing development tasks without human intervention.

Offers a unified LLM interface, agent runtime loop, terminal UI, and coding-agent CLI that reads, edits, and executes code to complete software development tasks.

Multi-agent LLM financial trading framework coordinating specialized agents for stock analysis, decision-making, and automated trading research for quantitative developers and researchers.

Autonomous research system where AI agents iteratively run training experiments to optimize language models overnight on single-GPU setups with minimal supervision.

Terminal-based coding agent optimized for open, low-cost models such as Kimi K3, providing Codex-like interaction and compatibility with multiple agent harnesses.

Provides an open-source AI penetration-testing team that dynamically executes code, verifies application vulnerabilities, and generates fix recommendations and compliance reports for security teams.

Research-oriented investment system that simulates an AI hedge-fund team, using cooperating agents for stock screening, strategy discussion, and portfolio backtesting.

Extensible open-source AI agent that runs locally via desktop app, CLI, and API, invoking tools and any compatible LLM to execute, edit, and test code and other tasks.

Self-hosted virtual companion container offering realtime voice chat and autonomous play in sandbox games like Minecraft and Factorio, available on web and desktop.

An AI pair programmer for your terminal that maps your codebase, edits files, commits to git, and fixes lint and test failures.

Ultra-lightweight self-hosted personal AI assistant runtime in about 4,000 lines of Python, supporting MCP, multiple chat channels, memory, skills, and scheduled automation.

Out-of-the-box super AI assistant and agent harness that plans tasks, uses tools and skills, and supports multi-agent, multi-model, multi-channel deployment with memory.

Terminal-based open-source coding agent written in Rust for reading projects, editing code, and running commands. Supports multiple models and collaborative agents to help developers complete programming tasks.

Extensible chatbot platform and development framework connecting multiple messaging apps with large language models and plugins. Lets operators deploy AI assistants with diverse features across IM services.

Local-first desktop personal AI assistant supporting customizable agents and workflows, MCP client integration, and local models via Ollama for private use.

Combines deterministic checks with an LLM agent that retrieves repository context and posts line-level review comments, allowing engineering teams to self-host it with their own models.

Provides reference agents and skills for financial-services teams, supporting end-to-end workflows such as roadshow notes, research, modeling, and reconciliation tasks.

Open-source AI coding assistant embedded in VS Code and other IDEs for chat, autocomplete, editing, and autonomous tasks. Supports multiple models and extensions for developers integrating AI into existing workflows.

Terminal-based coding agent for DeepSeek distributed as a single Go binary, featuring planning mode, sandboxing, and checkpoints for long-running autonomous programming locally.

Offers a self-hostable personal AI assistant deployable locally or in the cloud, connecting multiple chat applications with layered memory and an extensible plugin system for everyday users and developers.

Self-hosted open-source AI coding assistant offering code completion, chat, and knowledge-base question answering. Provides teams wanting a private Copilot alternative with control over models and data.

Gives individual traders a personal AI trading agent installable with one command, covering strategy research, backtesting, order execution, and market analysis across the full trading workflow.

A multi-agent AI developer that plans, codes, reviews, and debugs full production-ready apps step by step.

Delivers an open-source terminal coding agent that works with multiple model backends and supports tool calls, multi-agent collaboration and streaming interaction for developers.

Ships a ready-to-use coding agent deeply integrated with the IDE, offering multi-model support and a large built-in toolset for developers who want assisted editing without complex setup.

Provides a lightweight, self-hostable infrastructure for autonomous personal AI assistants that run on any OS, deploy anywhere and execute tasks across platforms.

Framework for building LLM-powered multi-agent applications more easily, with multi-model access, agent orchestration, deployable services, and a web UI for developers prototyping collaborative agents.

A Chinese multi-agent LLM framework for financial trading that coordinates specialized agents for market analysis, strategy evaluation, and investment decision support.

An open-source AI voice chatbot solution based on ESP32 supporting wake-word activation, spoken dialogue, and linkage control across multiple peripheral devices.

Ships a ready-to-use deep agent from LangChain with subagents, file access, persistent memory, and tools. Built on LangGraph, it helps developers create production-grade autonomous agents.

Runs a terminal-based AI coding agent supporting multiple models, session management, and LSP/MCP extensions. Built for developers, it assists with coding tasks and everyday development workflows across platforms.

Delivers an open-source terminal coding agent that understands repositories, edits files, and executes tasks. It supports multiple models and integrates with IDEs, desktops, and browsers.

Builds an autonomous agent for in-depth financial research that plans investigations, calls market-data tools, and self-validates its findings before answering complex investment questions.

Open-source multi-platform AI coding agent that works in editors and the terminal, supports multiple model providers, and autonomously completes coding tasks.

Fully local autonomous agent that browses the web, searches, writes and debugs multilingual code, executes programs, and plans complex tasks without cloud dependence.

Fullscreen mouse-interactive terminal harness and interface for coding agents, helping developers understand repositories, edit files, and manage long-running implementation tasks.

Provides an open-source AI-assisted design tool for creating and editing React interfaces. Designers generate or modify UI through conversational prompts combined with direct visual editing.

Releases open-source smartphone agent models and framework including AutoGLM-Phone-9B weights. The complete system enables autonomous control of Android devices to accomplish user tasks.

System connecting LLMs with Hugging Face models, where an LLM plans and orchestrates multiple specialized models to collaboratively complete complex AI tasks for researchers.

A fully autonomous AI penetration-testing system that coordinates multiple specialized agents to perform reconnaissance, exploitation, and report generation for security testing workflows.

A platform for stateful agents with long-term memory that runs in terminals, desktops, and clouds, allowing developers to build adaptive systems that learn and evolve over time.

Independently buildable and debuggable distribution of a Claude Code-like coding agent, focused on production-grade engineering, reliability, and memory-leak fixes for teams.

Agent-native automated trading platform aggregating multi-market data, where collaborating trading agents generate signals, manage portfolios, and execute copy-trading orders autonomously.

Local open-source AI app builder letting users supply their own API keys and generate complete cross-platform applications through conversational prompts without vendor lock-in.

Self-improving coding agent for long-running autonomous workflows, persisting in background while writing code and spawning subagents for teams handling complex software tasks.

Autonomous coding agent that resolves GitHub issues by using tools to edit code, fix vulnerabilities, and complete customizable software engineering tasks.

Open-source AI management system that dispatches dedicated agents in isolated cloud sandboxes to complete tasks such as reports and code autonomously, delivering results after human review.

Multi-agent general assistant built on the CAMEL framework, coordinating specialized agents and external tools to automatically complete diverse real-world tasks for end users.

Provides a lightweight coding-agent CLI and terminal interface for end users, supporting multi-session programming tasks with minimal memory footprint and rapid startup similar to Claude Code.

Open-source data assistant connecting databases, files, and warehouses, translating natural language into SQL, analysis, charts, and reports while supporting custom data-agent application development.

Open-source AI software engineer that interprets high-level instructions, plans implementation, researches online, and writes code in multiple languages to complete complex development tasks.

Open-source general-purpose assistant that gives agents a complete Linux computer, controlling desktop browsers and documents while orchestrating sub-agents to complete diverse tasks.

Rust-based open-source agent operating system shipped as a single binary with seven built-in scheduled autonomous agents for research, outreach, and sentiment tasks.

Desktop-resident open-source AI colleague that works across files to autonomously deliver code reviews, documentation drafts, and event analysis for developers and small teams.

Desktop AI teammate for knowledge workers that indexes email and meetings into a memory knowledge graph, working autonomously across mail, notes, browser, and code.

Open-source personal AI assistant runnable locally that combines tools, context, memory, and agent execution to complete everyday tasks autonomously for users.

OpenClaw-based multi-agent collaboration system modeled on Tang-dynasty Three Departments, using twelve roles for planning, review, and execution with live dashboard and audit.

Token-efficient image-to-3D tool that rebuilds objects from reference images as code-only procedural Three.js models, enforcing quality gates and producing animation-ready assets for web developers.

Terminal-based AI coding engine for developers handling complex tasks, planning and executing large multi-file changes with very large context windows and automated debugging support.

LLM-powered automated penetration testing agent supporting CTF solving and real-world engagements with multi-stage autonomous execution, session persistence, and multi-model backends for security professionals.

AI agent for automating data-driven R&D across quant, data science, and Kaggle scenarios, autonomously proposing ideas and continuously implementing iterative improvements.

Self-evolving autonomous agent grown from a compact seed with a dynamic skill tree, using minimal atomic tools to control browsers, terminals, files, keyboards, and mice with low token consumption.

Acts as a terminal-native AI coding assistant that reads and writes code, executes commands, manages Git operations, and uses persistent memory to maintain project understanding for developers across sessions.

A modular, low-latency pipeline for building local voice agents from open-source speech recognition, language, and synthesis models. Supports developers creating conversational and robotic dialogue applications fully on-device.

Open-source AI trading terminal for US stocks, commodities, forex, and crypto, where language models monitor markets and place orders under risk controls, with live multi-strategy ranking.

Open-source computer-use agent that operates real desktop applications with mouse, keyboard, and screen perception like a human, achieving top OSWorld scores for automation researchers.

the first library to let you embed a developer agent in your own app!

LLM coding agent for general software engineering offering CLI interaction, understanding natural language and orchestrating multiple tools and models to complete development workflows autonomously.

Open-source AI quantitative trading platform covering multi-agent research, strategy development, backtesting, paper trading, live execution, and monitoring for crypto, stocks, and forex.

An open-source AI trading operating system and multi-tenant SaaS platform for researching markets, building Python strategies, backtesting, paper and live trading across crypto, stocks, and forex for operators.

Community-driven multi-agent financial platform offering stock screening, in-depth research, news tracking, and automated trading agents for individual investors and quantitative researchers.

A self-hosted, offline, ChatGPT-like chatbot. Powered by Llama 2. 100% private, with no data leaving your device. New: Code Llama support!

Developed as an open-source desktop AI coworker for general office tasks, with the repository now archived and no longer maintained for prospective users and contributors.

Open-source asynchronous coding agent factory that picks up tasks, explores repositories in isolated sandboxes, validates changes, opens pull requests, and monitors reviews and CI.

An open-source multi-provider terminal coding agent built in Rust, offering a rich TUI pair-programming experience with plugins and editor integration.

MCP Server for Ghidra

Personal AI assistant system prioritizes on-device execution with memory, scheduled tasks, evaluation, and cloud fallback, designed for individuals wanting private automation across personal devices.

An open-source mobile agent that controls Android and iOS devices through natural language, combining UI-aware screenshot understanding with multi-step automation via CLI or Python.

AI customer-service system for Xianyu sellers using multi-expert collaboration, intelligent bargaining, and contextual dialogue to enable around-the-clock automated reception on the marketplace.

Lightweight 7MB terminal-first AI-native development workspace integrating terminal, code editing, and an autonomous AI sidebar for developers working in agent-assisted coding environments.

Provides an open-source AI-native design editor compatible with Figma files for designers and developers, supporting AI-assisted design generation, real-time collaboration, and export from designs to code.

Delivers an LLM-driven general-purpose autonomous agent that plans tasks, invokes external tools, and solves complex workflows, supporting human collaboration and graphical interaction for interactive task execution.

Open-source multi-agent platform for finance that combines LLMs with quantitative analysis to support automated investment research, trading strategy development and risk assessment.

Security scanning harness that uses coding agents to find vulnerabilities across large codebases through autonomous parallel code review and produces readable reports for developers.

Generates executable tasks from conversations and runs them autonomously across devices, offering isolated workspaces, accumulated knowledge management, and open API integrations for developers building personal automation.

Native macOS harness for personal AI agents, running any model fully offline with persistent memory, autonomous execution, cryptographic identity, and Swift-based local control.

Local-first open-source Claude Design alternative that turns prompts into prototypes, slides, and PDFs using BYOK multi-model support for designers and developers.

A minimalist hundred-line AI software engineering agent that reads code, runs shell commands, and fixes GitHub issues while scoring strongly on SWE-bench Verified.

Terminal-based AI coding agent that reads and edits code, executes commands, and plans subsequent steps autonomously during software tasks for developers.

ChatLaw：A Powerful LLM Tailored for Chinese Legal. 中文法律大模型

Six-month live experiment where ChatGPT manages a real-money micro-cap portfolio, including a trading engine, position logs and evaluation reports as a baseline framework for LLM investment research.

AI一键批量生成各类短视频,自动批量混剪短视频,自动把视频发布到抖音,快手,小红书,视频号上,赚钱从来没有这么容易过! 支持本地语音模型chatTTS,fasterwhisper,GPTSoVITS,支持云语音：Azure,阿里云,腾讯云。支持Stable diffusion,comfyUI直接AI生图。Generate short videos with one click using AI LLM,print money together! support:chatTTS,faster-whisper,GPTSoVITS,Azure,tencent Cloud,Ali Cloud.

Browser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser

An open-source multi-agent system for source-code vulnerability discovery that supports one-click deployment, collaborative auditing, sandboxed PoC verification, and report generation.

Offers a token-efficient general-purpose AI assistant with routable memory and multiple entry points, helping developers run capable agents at lower cost without sacrificing intelligence density.

Provides an AI-native security testing platform in Go for penetration testers, integrating 100+ tools with intelligent orchestration, role-based testing, skills, and lifecycle management.

Open-source AI product-design agent living inside the IDE, generating interfaces and component code autonomously from natural-language descriptions for designers and developers.

🔍 An LLM-based Multi-agent Framework of Web Search Engine (like Perplexity.ai Pro and SearchGPT)

AppAgent: Multimodal Agents as Smartphone Users, an LLM-based multimodal agent framework designed to operate smartphone apps.

Personal AI agent with persistent memory, distinct personality, and autonomous capabilities, designed for everyday conversations, trivial tasks, and long-term companionship.