Havuz

NeoCode 是一个本地优先的 AI Coding Agent，支持终端、Web、桌面端与飞书等多入口接入，帮助用户读懂项目、修改代码、运行命令并安全推进开发任务

🤖 Bony Agent — Full-stack AI content production & distribution platform. One person can run an entire team: AI chat · image/video/music gene

AI coding agent that shows its work — every step visible, every patch approved by you. Build projects brick by brick, stay in control.

A group chat AI character that learns from feedback.

Sable is a self-hosted agentic platform that lives on your machine and grows with you. Part AI companion, part dev environment, part chaos e

QQLore 是一个面向IM聊天记录的本地智能分析与检索系统，基于 TypeScript / Node.js 构建，覆盖聊天记录读取与增量同步、消息清洗、上下文重建、语义向量化、RAG 检索、Agent 问答和周期性摘要等完整链路。系统支持多群组管理、兴趣度分析与 WebUI 可

[EMNLP main 2026]Co-Evolving World Models and Agent Policies for LLM Agents

Cognitive-structured Multimodal Agent (CMA-Harness): a memory-centric agent for long-horizon multimodal understanding, generation, and editi

⚽️🤖 Benchmarking LLMs and deep-research agents on real-world football prediction — from the tactical "who scores in minute 67" to the strate

Umbrella for the LLM Dark Patterns Hooks suite — single-purpose Claude Code Stop hooks that suppress sycophancy, paternalism, false-success,

Hi Code — 本地优先的桌面编码 Agent 工作台 · a local-first desktop coding agent workbench for any OpenAI-compatible model

Turn Claude Code into a voice-enabled, multi-channel assistant that runs tasks in parallel

ZW3D CAD automation bridge and iterative drafting agent for mechanical drawing generation.

Talk to your home. A voice-driven personal AI agent that chats, assists, and orchestrates your smart home.

A human-like memory for AI Agent

Windows computer-use MCP server: drive any desktop app via semantic discover-then-act targeting (entities + leases, not pixel coordinates),

The job hunt, in one window. With your agent. An open-source, agent-native, local-first desktop workspace for job discovery, browsing, docum

🌏Earth Online🌏：AI-powered travel memory engine that automatically groups photos into trips and visualizes them on an interactive map

Unity-Open-MCP is a complete set of tools to use AI with Unity. It acts as a bridge between AI agents (like Claude Code, Cursor, ZCode) and

CLI for recommending and managing agent skills from detected project dependencies, with curated mappings for common libraries.

Rust port of earendil-works/pi — coding agent harness (pi-ai + pi-agent + pi CLI)

Build & host live AI agents with 5 lines of code.

Real-world browser-agent benchmark: 210 tasks across 107 websites, multi-agent/multi-browser evaluation, reproducible leaderboard and result

A more autonomous, more persistent, and easier alternative to OpenClaw/Hermes. Give it a task and it will carry it out, even if it takes

🔥PaiCLI — Terminal AI Agent in TypeScript

ByteMind is a lightweight Coding Agent designed to help users complete the software engineering loop.

20位天涯大神的多智能体决策系统，multi-agent team

A declarative, hermetic harness for Nix-defined agents.

See the full type-inference chain of any value in PHPStan, not just a single snapshot.

Multi-agent LLM pipeline for long-form novel continuation · 多 Agent 长篇小说续写流水线，19 轮迭代 + 全程 mock-first

Grok Build plugin: hand work to local Claude Code or Codex as a background task

⚡ A floating, agentic DeepSeek AI assistant that overlays every app on Android — clipboard quick-actions + a chat that can run shell command

Drive a GNOME Wayland desktop from AI agents. CLI-first, no portals, no consent dialogs.

Pure-Rust async runtime for LLM agents. Multi-provider (Anthropic/OpenAI/Ollama), tool-capable end-to-end, structured permissions, real MCP

Factorly is a local runtime for agent tool chains. It manages tool calls, injects credentials from an encrypted vault, enforces governance r

The board your Hermes agent builds with you — live widgets, sandboxed custom widgets, and operator-governed external tools, for the web dash

Agent skills and plugins to give your agents access to Pruna API and generation workflows.

Agent-friendly Markdown-to-PDF CLI with local PDF rendering and JSON diagnostics.

Local-first, auditable memory for AI agents, with governed retrieval, delegated context delivery, and verifiable execution evidence. Works w

Your personal AI workbench on desktop. Execute tools, install skills, connect MCP servers, build persistent memory — all models, one interfa

Mint 是本地优先的 AI 知识工作台：将文档、网页与对话沉淀为可持续积累的 LLM Wiki，借助 Agent 与记忆 理解和复用个人知识。采用 SDD 规格驱动开发，以自动化验收闭环交付。

Agent registration, discovery, and reliable async messaging — in one broker.

Agent Harness or sum

Agent Doris CLI

CLI to organize temporary agent knowledge into scratchpads, with a read-only visual viewer.

Build AI agents using the same architecture patterns as Claude Code. Skill + 6 examples + runnable Python code. No framework.

Java-implemented extensible AI Agent runtime — direct Anthropic protocol connection / Spring Boot 4 assembly / multi-tool collaboration / We

OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language World Models

一个基于 LangChain + FastAPI 的扫地机器人领域智能客服 Agent，支持 RAG 知识库问答、多轮对话、JWT 认证与会话隔离、Redis 缓存与限流。适用于初学者自学RAG和Agent项目

加密环境文件操作 MCP 工具。当 Node.js 是加密软件白名单进程时，通过 fs 模块自动解密读写文件明文，替代 AI Agent 内置文件工具，适用于任何支持 MCP 协议的 Agent。