编码 Agent

SWE-AF 是一个自主软件工程舰队平台，采用多 Agent 工厂架构，通过调度器、编码 Agent、审查 Agent 和验证 Agent 的协同工作，实现从问题分析到代码修复的全自动化软件工程流程，在基准测试中获得 95/100 分。

终端中的 AI 结对编程工具，支持与主流 LLM 协作进行代码编辑、Git 管理和多文件重构，深度集成开发者工作流。

AutoCodeRover is a project structure-aware autonomous software engineer agent that achieves automated program repair and issue resolution by understanding the overall codebase architecture.

An AI agent that writes actually useful code for you by writing tests first, then generating code to pass them.

An open-source background agents coding system that autonomously executes coding tasks in the background, including code reviews, test generation, and feature implementation.

面向并行编码 Agent 的编排工具，自动规划任务、生成 Agent 实例并自主处理 CI 修复、合并冲突和代码审查，支持多 Agent 协同完成复杂开发任务。

Turn Claude Code into a full game dev studio with 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

gptme 是一个终端 AI Agent 工具，支持代码编写、终端操作和网页浏览，可在本地构建持久化的自主 Agent。

完全本地化的 Manus AI 替代方案，支持自主浏览网页、编写代码和语音交互，无需任何 API 费用

开源编程 Agent CLI，支持 OpenAI、Gemini、DeepSeek、Ollama、Codex 及 200+ 模型的 OpenAI 兼容 API，是 Claude Code 的开源替代方案。

Ship AI Agents to Google Cloud in minutes, not months. Production-ready templates with built-in CI/CD, evaluation, and observability.

DeepCode is an open agentic coding platform supporting Paper2Code, Text2Web, and Text2Backend, leveraging agent technology for automated software development workflows.

Git LRC is a free, unlimited AI code review tool that runs automatically on every commit, helping developers catch and fix code issues early in the development workflow.

A set of ready-to-use Agent Skills for research, science, engineering, analysis, finance and writing across multiple coding agents.

KiloCode 是一体化开源编程 Agent 平台，支持 VS Code 和 JetBrains，集成 200+ 模型，提供自主编码、调试和迭代能力，是 OpenRouter 上使用量最大的编程 Agent。

Kodezi Chronos is a debugging-first language model achieving state-of-the-art performance on SWE-bench, capable of autonomously handling software debugging and code repair tasks.

用Rust编写的终端编程Agent，提供高性能和内存安全。支持代码生成、文件编辑、命令执行等完整的开发工作流。

A tool for managing project collaboration between humans and AI Agents in a git ecosystem.

Locally runnable version of Claude Code with cross-platform desktop app and Computer Use capabilities, including core module analysis.

Warcraft III Peon voice notifications for Claude Code, Codex, IDEs, and any AI agent — stop babysitting your terminal, employ a Peon today.

An open-source AI coding agent that lives in your terminal, built by Qwen team with support for code generation, editing, debugging and multi-file operations.

SWE-agent 能自动分析 GitHub Issue 并用 LLM 生成修复代码，支持网络安全审计和编程竞赛场景，NeurIPS 2024 论文项目。

Mini SWE-Agent is a minimalist AI agent in just 100 lines of code that solves GitHub issues or assists developers in the command line, demonstrating core coding agent capabilities with minimal implementation.

SWE-bench is a benchmark for evaluating language models on real-world GitHub issue resolution, featuring genuine problems from popular Python repositories, now a core standard for measuring AI coding agent capabilities.

SWE Agent 训练数据规模化生成工具包（NeurIPS 2025 D&B Spotlight），用于自动创建大规模软件工程 Agent 训练数据集，推动 SWE Agent 能力提升。

Tabby 是一个可自托管的 AI 编程助手，支持代码补全、代码生成和企业级部署，兼容主流 IDE。

开源 AI PR 审查 Agent，自动分析 Pull Request 并生成代码审查意见、改进建议和 PR 描述，支持 GitHub、GitLab、Bitbucket 等主流代码托管平台。

Zero-Config Code Flow for Claude Code and Codex, providing one-click project initialization and context management.

开源可扩展的 AI 编程 Agent，超越代码补全，支持安装、执行、编辑和测试全流程，兼容多种 LLM 后端。

Agent Skills 是由 Addy Osmani 维护的生产级工程技能集合，为 AI 编码智能体提供经过实战验证的工程最佳实践与操作规范。

The agent harness performance optimization system with skills, instincts, memory, security and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

开源编程Agent，在终端中运行的AI编程助手。支持代码生成、重构、调试、测试等全流程开发任务，提供智能代码补全和建议。

Claude Code 是 Anthropic 推出的终端 AI 编程助手，能理解代码库、执行任务、解释复杂代码并处理 Git 工作流。

Official Anthropic repository for Agent Skills, providing ready-to-use Claude agent skill examples and templates.

GPT Engineer 是一个根据自然语言描述生成整个代码库的 AI 工具。你只需描述想要构建的内容，AI 会提出澄清问题，然后生成完整的代码。

Terminal session manager for AI coding agents. One TUI for Claude, Gemini, OpenCode, Codex, and more, enabling session switching and parallel workflows.

A lightweight, cross-platform code editor built with Tauri (Rust and React) with Git support, AI agents, and vim keybindings.

Augment SWE-bench Agent is the number one open-source SWE-bench Verified implementation, demonstrating how to build high-performance software engineering agents to automatically resolve GitHub issues.

Multi-functional interface for GPT/GLM LLMs with optimized academic paper reading, polishing and writing. Supports multi-model parallel, plugin extensions and local deployment.

ByteRover CLI 是一个为自主编码 Agent 提供持久化结构化记忆的便携工具，支持上下文树管理、类 Git 版本控制、云端同步，兼容 Cursor、Claude Code、Windsurf 等 22+ 编码 Agent，集成 MCP 协议。

The toolkit for AI devtools context engineering. Build with codebase mapping, symbol extraction, and many kinds of code search to help AI agents better understand and operate on codebases.

由 Charm 团队打造的开源智能编码 CLI，支持多种 LLM 后端，提供终端内的自主编程体验

开源 AI 原生终端，专为云和基础设施管理设计，支持用自然语言部署、排查和自动化运维服务，集成智能 Agent 能力。

Cline 是一款运行在 IDE 中的自主编码 Agent，能够创建/编辑文件、执行命令、使用浏览器，并在每一步请求用户确认。

An MCP server and CLI tool that indexes local code into a graph database to provide precise code context for AI assistants.

An introductory guide to context engineering - the systematic approach to building high-quality context for AI coding assistants, centered on Claude Code but applicable to any AI coding tool.

Continue 是一个开源的 AI 代码助手扩展，支持 VS Code 和 JetBrains IDE。可以自动补全代码、重构、解释代码，帮助开发者提高编程效率。

Dyad 是一款本地运行的 AI 应用构建工具，让用户通过自然语言对话快速创建和迭代 Web 应用，支持多种 LLM 提供商。

Python 和 JS/TS SDK，用于在安全的云端沙箱中运行 AI 生成的代码，支持 Jupyter Notebook 风格的代码解释

A cross-platform desktop All-in-One assistant tool for managing Claude Code, Codex, OpenCode, OpenClaw and Gemini CLI agents in one place.

OpenSpec is a spec-driven development (SDD) platform that guides AI coding assistants to generate code through specification definitions, improving development efficiency and code quality.

Agent Zero 是一个通用 AI Agent 框架，支持自主任务规划、工具使用和代码执行，可用于构建各类自主 AI 助手。

Garry Tan's opinionated OpenClaw/Hermes Agent Brain with optimized system prompts and agent workflows.

编程 Agent 的记忆增强工具，为 Claude Code、Codex 等编码 Agent 提供持久化上下文记忆，大幅提升长任务的一致性。

开源智能开发环境（YC W26），支持并行运行多个编码 Agent，兼容任意 LLM 提供商

GitHub's official community-contributed collection of instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.

GitHub官方Copilot命令行工具，将Copilot编程Agent的能力直接带到终端。支持代码生成、命令建议、错误修复等。

Gemini CLI 是 Google 推出的终端 AI Agent 工具，支持代码生成、文件操作和多轮对话，提供免费使用层级。

用自然语言描述软件需求，让 AI 自动生成并执行代码的实验平台，支持 Vision 模型输入和开源模型接入

A light-weight and powerful meta-prompting, context engineering, and spec-driven development system for AI coding agents like Claude Code.

HumanLayer 为 AI 编程 Agent 提供人机协作层，让 Agent 在复杂代码库中遇到难题时能寻求人类审批和指导，提升代码生成准确性与安全性。

统一的 AI 编程 Agent 规则管理工具，可将相同的编码规则和配置一键应用到 Claude Code、Cursor、Copilot 等所有主流编程 Agent。

Xcode 首个 AI 编程助手扩展，支持 GitHub Copilot、Codeium 和 ChatGPT，将主流 AI 代码补全和对话能力直接集成到 Apple 开发工作流。

由 Andrej Karpathy 开发的 AI 研究自动化 Agent，能在单 GPU 上自动运行纳米聊天训练研究实验。

Claude Coder 是一款 VSCode 扩展形式的自主编码 Agent，能将设计稿转化为代码、自动修复 lint 错误、编写测试并执行复杂的多文件代码变更，支持 Agent 模式进行自主任务执行。

Kodus AI is an open-source AI code review tool with full control over model choice and costs, automatically analyzing pull requests and delivering high-quality code review feedback.

DevOpsGPT is a multi-agent system for AI-driven software development that combines LLMs with DevOps tools to convert natural language requirements into working software, supporting any development language and extending existing codebases.

LangChain 开源的异步编码 Agent，基于 LangGraph 构建，支持自主完成软件工程任务，包括代码生成、调试和文件编辑等异步工作流。

AI-powered software engineering multi-agent system with researcher and developer agents that automate code implementation through intelligent planning and execution.

Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents. Manage multiple coding agent sessions efficiently.

A CLI for Git worktree management, designed for parallel AI agent workflows. Run multiple AI coding agents simultaneously across branches.

12 Lessons to Get Started Building AI Agents by Microsoft. Hands-on curriculum covering core agent concepts, tool use, and multi-agent collaboration.

Context Mode 是面向 AI 编程 Agent 的上下文窗口优化工具，通过沙盒化工具输出实现 98% 的上下文缩减，支持 12 个主流编程平台。

多语言软件工程 Agent 基准测试集，扩展 SWE-bench 至多语言场景，用于评估 AI Agent 跨编程语言的 Issue 解决能力。

Multica 是开源的托管式 Agent 平台，将编程 Agent 转化为真正的团队成员，支持任务分配、进度追踪和技能积累，实现 Agent 能力的持续复合增长。

Aide 是一款 VSCode AI 编程辅助扩展，提供一键注释、代码转换、UI 生成代码和 AI 批量文件处理等功能。

NousResearch \u63A8\u51FA\u7684\u81EA\u4E3B AI Agent \u6846\u67B6\uFF0C\u652F\ \u6301\u591A\u79CD LLM \u540E\u7AEF\uFF0C\u80FD\u591F\u968F\u7528\u6237\u9700\ \u6C42\u4E0D\u65AD\u6210\u957F\u548C\u9002\u5E94\u3002

The Monorepo Platform that amplifies both developers and AI agents, optimizing builds, scaling CI and automatically fixing failed PRs.

An agentic skills framework and software development methodology that provides reusable skill modules and engineered workflows for AI coding agents.

Multi-model AI agent desktop client that connects to any AI provider, extends with MCP and skills, and supports remote control from your phone. Built with Electron and Next.js.

SWE-Lancer is an OpenAI benchmark dataset evaluating frontier language models on freelance software engineering tasks, covering real scenarios from simple bug fixes to complex feature development.

Codex CLI 是 OpenAI 开源的代码 Agent 命令行工具，面向代码理解、重构、生成与终端协作场景，适合作为开发者工作流中的编程 Agent 接口。

OpenAI 开源的编码 Agent 编排工具，将项目工作转化为隔离的自主实现运行，让团队能够管理任务而非监督编码 Agent 的每一步操作。

强大的AI编程Agent，专为终端设计。支持代码生成、重构、调试，提供智能建议和自动化工作流。

A framework for quickly building AI-native IDE products with built-in MCP client support for deep integration of coding assistants.

强大的 MCP 编码工具包，提供语义化代码检索和编辑能力，堪称 AI Agent 的 IDE

Aider 是一个 AI 编程助手，可以在终端中与 LLM 协作编程。支持 Git 集成，可以自动提交代码更改，适合开发者进行结对编程。

AI原生多Agent开发平台，基于Kotlin Multiplatform构建，覆盖软件开发全生命周期的7个阶段。支持自动代码生成、测试、部署、文档编写等全流程开发任务。

开源 AI 编码 Agent，专为大型项目和真实开发任务设计，支持终端操作、文件管理和多步骤代码生成，适合复杂软件工程场景。

AI-friendly semantic code search engine for large codebases — combines ripgrep speed with tree-sitter AST parsing for precise, context-aware code understanding in AI coding assistants.

Quantalogic is a ReAct-based coding agent framework supporting multiple LLM backends, with tool use, reasoning chain management, and an extensible plugin system.

Roo Code 是一款运行在 VS Code 和 JetBrains 中的自主编码 Agent 扩展，能在编辑器中直接创建/编辑文件和执行终端命令。

AI-powered job search system built on Claude Code with 14 skill modes, Go dashboard, PDF generation and batch processing.

Python scraper based on AI that uses LLMs and knowledge graphs to automatically build web data extraction pipelines.

From vibe coding to agentic engineering — a practice guide helping developers master Claude Code best practices and advanced techniques.

A nano claude code-like agent harness built from scratch, demonstrating how to build AI coding assistants from zero to one.

开源的 Claude Code / Cursor CLI / Codex Web 界面，支持在移动端和 Web 端远程管理 AI 编码会话和项目。

Skales 是一个本地运行的 AI 桌面 Agent，支持 Windows、macOS 和 Linux。提供 Agent 技能系统（SKILL.md）、自主编码（Codework）、多 Agent 团队协作和桌面自动化，无需 Docker 或终端，支持 15+ AI 提供商。

Refact is a Rust-based AI coding agent that handles engineering tasks end-to-end, integrating into developer workflows with code completion, chat, agent actions, and self-hosted deployment.

多 AI 终端代理管理器，可在终端中同时运行和管理 Claude Code、Codex、OpenCode 等多个编码代理

An autonomous AI agent loop that runs repeatedly until all PRD items are complete, automating the entire development cycle.

Sourcery is an instant AI code review tool that automatically detects code issues, suggests refactoring, and improves code quality, integrating into developer workflows for real-time code review.

OpenCode 是一个开源终端编码 Agent，支持多种 LLM 提供商，在终端中提供 AI 驱动的代码生成和编辑能力。

stagewise 是专为 Web 开发者构建的浏览器内编程 Agent，将 AI 编码能力直接集成到浏览器中，支持从网页界面直接与代码交互。

首个开源 Agentic 软件工程师，对标 Devin，支持 Claude 3、GPT-4、Gemini 等多模型，具备规划推理和网络浏览能力

An open-source autonomous agent powered by Grok, capable of executing tasks, browsing the web, and generating code directly from your terminal.

面向 AI Agent 时代的代码编辑器，支持并行运行多个 Claude Code、Codex 等编码 Agent 实例

Sweep is an AI coding assistant for JetBrains IDEs that automatically resolves GitHub issues and submits code changes, automating software development workflows.

A Claude Code plugin that automatically captures coding session context, compresses it with AI, and injects relevant context back into future sessions for persistent memory.

Open source autonomous software development system that automates the entire process from requirements to code using LLMs.

A curated directory of open-source AI agent skills for Swift and Apple platform development.

OpenReview is an open-source, self-hosted AI code review bot powered by Vercel that automatically analyzes pull requests and provides code review suggestions.

Void 是一款开源的 AI 代码编辑器，基于 VS Code 架构，支持 Claude、GPT 等多种模型，提供 Cursor 风格的智能编程体验。

A powerful GUI app and Toolkit for Claude Code — create custom agents, manage interactive Claude Code sessions, run secure background agents, and more.

Repomix 可将整个代码仓库打包为单一 AI 友好文件，方便将代码库提供给 Claude、ChatGPT、DeepSeek 等 LLM 进行分析、审查或生成代码。

Neovim AI 编程插件，将 Cursor AI IDE 的体验带入 Neovim，支持 AI 驱动的代码生成、编辑和聊天，深度集成 Neovim 生态。

Zed Agentic 是 Zed 围绕编辑器内 Agent 协作打造的开源项目，强调代码理解、编辑建议与开发者工作流增强，适合关注 IDE 内编程 Agent 的场景。

Strict AI coder for enterprises, quality first, including AI Agent, AI CodeReview, and AI Completion.

Code search MCP for Claude Code and coding agents. Makes entire codebases available as context for AI coding assistants using vector-based semantic code search for precise understanding of large projects.