AI工具

多功能工具箱 · 桌面端开源版 | 音视频/图片/PDF/AI 一站式处理

FireRedTTS3: Multilingual and Multi-Dialect Voice Cloning with Instruction-Guided Voice Design and Speech Editing

Visibility into AI agent activity on endpoints, with on-device detection, optional pre-action blocking, and forensic reconstruction.

Patter是一个开源SDK，为AI代理提供电话能力，处理语音识别、合成、实时对话及电话运营商集成。

Echo Agent 是一个可自托管、长期运行、持续学习的 AI Agent，面向个人与团队的私有自动化场景。它可以部署在自有服务器上，统一连接模型、工具、记忆、权限与消息入口。内置四层认知记忆、遗忘曲线与矛盾检测机制，能够在跨会话任务中持续沉淀上下文，并保持长期记忆的质量。针对命令执行、文件操作等高风险行为，它提供基于 LLM 的审批与解释机制，为关键操作建立可审计、可追溯的安全边界。原生支持 MCP、A2A、多模型路由、任务调度、工具调用和多通道接入，覆盖 CLI、Gateway API、微信、Telegram 等入口。它让 Agent 带着长期记忆和可进化技能，持续、安全地为你工作。

RTX 6000 Pro Wiki - Running Large LLMs (Qwen3.5-397B, Kimi-K2.5, GLM-5) on PCIe GPUs without NVLink

Build agents which are controlled by LLMs

An AI copywriter that uses real copywriting skills + real marketing knowledge with human tone.

Your Open Autonomous Android Agent - A production-ready, self-planning AI assistant powered by local/remote LLMs and accessibility-driven screen automation.

Autonomous GPT-4 agent platform

在Android设备上运行本地AI模型实现手机自动化的开源应用

An LLM-powered repository agent designed to assist developers and teams in generating documentation and understanding repositories quickly

An ops AI Agent that understands your infrastructure, finds the root cause, and fixes it - right from Slack, Telegram, Lark or DingTalk.

All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine

开源数字永生框架，通过四维蒸馏引擎将人的认知从聊天记录、文档中结构化提取为AI可加载的数字分身。

整合 Claude Code 编程能力与 Hermes Agent 自进化能力的智能体 / An AI agent combining Claude Code coding capabilities with Hermes Agent self-evolution.

AI Image Prompts - 10,000+ curated prompts for any model. Works with Nano Banana Pro, Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E, Flux, Stable Diffusion, and more.

Multimodal RL training framework for diffusion & omni models

WHartTest 是一款AI驱动的测试自动化平台，实现从需求到可执行测试用例的自动化生成与管理，帮助测试团队提升效率与覆盖率。 (WHartTest is an AI-driven test automation platform that automates the generation and management of executable test cases from requirements, helping testing teams improve efficiency and coverage.)

Evidence-driven Agent Skills for design award research, evaluation, award matching, entry writing, and submission readiness.

Production-grade C++ edge AI engine for video analytics with visual pipeline orchestration, on-device VLM, and real-time OSD.

ExploitGym is a large-scale, realistic benchmark built from real-world vulnerabilities designed to evaluate AI agents' ability to develop exploits.

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary. Self-host or use managed cloud.

Web file manager with uploads, tagging, share links, gallery/table views, and an in-browser editor. ([Demo](https://github.com/error311/FileRise?tab=readme-ov-file#live-demo)) `MIT` `Docker/PHP`

一个开源的文件系统和知识库AI代理模板，支持无需嵌入向量的文件搜索，可多平台部署。

[Local Deep Research](https://github.com/LearningCircuit/local-deep-research) - A deep research tool for searching academic sources, the web, and private documents with local or cloud LLMs. [#opensource](https://github.com/LearningCircuit/local-deep-research)

AI-powered multi-voice audiobook generator - LLM script annotation, voice cloning, voice design, LoRA training, per-line style control, and export to MP3, chaptered M4B, or Audacity multi-track. Built on Qwen3-TTS.

Qwen-AgentWorld是一个原生语言世界模型，用于模拟智能体环境并在七个统一领域（MCP、搜索、终端、SWE、Android、Web、OS）中通过长链思维推理进行环境建模和评估。

Agentic AI security tool that applies proactive, attacker-first analysis directly to source code.

Game-development Agent Skills for AI coding agents: install once and a master router loads the right skill for your engine and task. 66 original, version-pinned skills (plus a master router) in the portable SKILL.md format that runs across Claude Code, Cursor, Codex, Copilot, Gemini CLI and more, for Godot, Unity, Unreal, web and beyond.

"AgentSpace: Human + Agents. One Team. One Workspace"

自托管的个人Agent工作台，支持通过飞书、Telegram、微信等渠道使用Codex和Kimi Code等AI引擎。

A web frontend for WeeChat. ([Demo](https://www.glowing-bear.org)) `GPL-3.0` `Nodejs`

Budgeting app based on the Bucket Budgeting Principle. `AGPL-3.0` `Docker/C#`

An autoagentic AGI that is self-evolving and modular.

Smart job hunting Agent - AI-powered automation from scraping to delivery

日本語UIをAIエージェントに正しくつくらせるためのDESIGN.md集。Japanese DESIGN.md collection for AI agents - extending Google Stitch format with CJK typography.

Hacking the Apple AirPort Time Capsule to run modern Samba

企业级云原生微服务基座，支持阻塞式和响应式服务并行，一套代码实现微服务和单体架构灵活切换

无需登录和API密钥即可从X/Twitter获取推文、回复、时间线、列表和文章的Python CLI工具，支持多后端自动回退。

基于AI的GitHub个人资料分析工具，提供评分、实时指标和可导出报告。

自监督视觉Transformer骨干网络系列，用于密集空间感知任务（深度估计、语义分割等）

A recipe keeper, meal plan organizer, and shopping list manager that can import recipes directly from any URL. ([Demo](https://recipesage.com)) `AGPL-3.0` `Nodejs`

Real-world AI penetration testing engineer for authorized assessments - built-in cloud module covering AWS/Azure/GCP + Aliyun/Tencent/Huawei clouds. Built on Claude Agent SDK

Complete trashmail solution that exposes an SMTP server and has a web interface to manage received emails. Works with multiple and wildcard domains and is fully file based (no database needed). Includes RSS feeds and JSON API. `Apache-2.0` `Python/PHP/Docker`

Voice-first local agent orchestration runtime for auditable DAG workflows.

The first open-source agentic AI physicist, by Physical Superintelligence PBC (PSI).

开源大规模MoE视频生成模型，用于具身智能，支持文本到视频、图像到视频等任务

List of molecular design using Generative AI and Deep Learning.

OpenCluely is a free, open source Cluely (alternative), built for technical interviews like DSA, OAs, and CP. It offers an invisible overlay, real-time AI help, Smart Image Processing for question capture, and multi-language support : 100% customizable and private.

开源肌腱驱动机器人手，提供完整硬件设计、3D打印文件、PCB、固件和Python SDK，用于灵巧操作研究。

A 10-week, 30-minutes-a-day roadmap for LLM inference serving and optimization. vLLM, SGLang, quantization, speculative decoding, benchmarking.

HTTP daemon for serving wikis from ZIM files. `GPL-3.0` `C++`

MAC (Multi-Agent CAD): A decoupled multi-agent framework for text-to-CAD generation via constrained test-time compute

Electron GUI app for the pi coding agent runtime

The all-in-one AI agent living in your terminal. Build landing, mobile apps, backends, manages files, deep research, schedule tasks and events. Self-improving, self-healing, fully autonomous. TUI, CLI and Daemon. Connect to your favorite channels.

Learn AI agents from scratch.

China's 1st enterprise multi-agent IT ops platform. LLM-powered auto-remediation for Zabbix/Prometheus. Docker deploy.

指向Claude Code源代码的指针仓库，实际代码已迁移至其他仓库，Claude Code是一个AI编码助手。

Secure MCP runtime for AI agents to operate local machines, servers, and containers with multi-device orchestration.

The spatial IDE for recursive multi-agent orchestration. It's like an Obsidian graph-view that you work directly inside of.

Open-source AI sandbox infrastructure with unified API for VMMs -- Firecracker, QEMU and libkrun.

技术小白也能拥有的 AI 第二大脑：无需翻墙、无需订阅，跟着图文 ~20 分钟装好 Obsidian + Claudian + cc-switch + DeepSeek V4。

VOZEB PRO：从创作简报、视觉方向、物料规划到一致性生成与自动视觉复盘的 AI 创作工作台。

Nintendo Switch library manager, with automated management tasks (file identification and organization, missing updates/DLC), serving your library to multiple supported clients on your Switch, with shop customization and multi user authentication. `AGPL-3.0` `Docker/Python`

Self-improving agentic QA harness for web and mobile tests. Write tests in natural language, use memory to adapt to UI changes, and catch regressions before releases ship.

一个基于AI的高考志愿填报顾问，内置24省官方录取数据库，提供智能分析、推荐和联网搜索功能。

Host your docs. Simple. Versioned. Fancy. `MIT` `Python/Docker`

面向 AI 影视创作的开源无限画布工作台，集成多模态生成、分镜编排、素材管理与 Agent 工作流。

为科研人员设计的自进化AI研究助手，具备自我进化技能、持久研究记忆、长时深度研究能力和零幻觉保障。

🎨 Open-source AI slide studio inside Codex: image-native decks, every slide a full visual canvas. ⚡ 10+ high-quality slides in ~4-5 minutes - Fast mode renders every page in parallel. 🔍 Watch the whole chain live: research → outline → style → render → edit → present → export PDF/PPTX. 🖥️ Browser-first · zero API keys · durable projects.

一个为AI应用和智能体设计的自进化记忆与上下文引擎，能够从对话、文档、工具调用等中提取结构化记忆，并通过多种接口提供给AI系统。

Simple and powerful finance tracker. ([Demo](https://wygiwyh-demo.herculino.com/)) `AGPL-3.0` `Docker/Python`

Windows Agent Arena是一个可扩展的Windows AI agent平台，用于测试和基准测试多模态桌面AI agent。

An evaluation and evolution tool for Agent Skills.

Self-improving AI that runs your whole business turning LLM tokens into work and dollars.

Every coding-agent session on your Mac in one place - browse, search, resume. Rust + GPUI.

A collection of reusable skills for AI coding agents - prompts, slash commands, and tools built for Unity workflows