OpenAgents: 在共用的開源工作區中將人類團隊與自主 AI 代理連結起來,但缺乏深入的本機桌面整合。 OpenAgents 每月有 38,286 次造訪,月增率 +48.4%,定價模式為免費增值。OpenBot 每月有 145,308 次造訪,成長率 +2,617.1%,定價模式為 Freemium, Free, Contact for Pricing。

OpenAgents is an open-source platform for building, connecting and orchestrating AI agent workflows inside a shared team workspace. It pairs a browser-based Workspace with a desktop Launcher so agents running on your own machine or a remote server can join the same threads, files and tasks as your teammates.
| 面向/指標 | ![]() | ![]() |
|---|---|---|
| 官方網域 | openagents.org | openbot.run |
| 定價模式 | 免費增值 | Freemium, Free, Contact for Pricing |
| 每月造訪次數 | 3.83萬 (38,286) | 14.53萬 (145,308) |
| 月流量變化量 | +1.25萬 (+48.4%) | +14萬 (+2,617.1%) |
| 主要市場與市佔 | 🇮🇳 印度 (26.2%) | 🇧🇷 巴西 (41.1%) |
| 每月直接造訪 | 1.17萬 | 6.28萬 |
| 使用者評分 | 4.5 | 4.4 |
| 支援平台 | 網站 | 網站 |

OpenAgents is an open-source platform for building, connecting and orchestrating AI agent workflows inside a shared team workspace. It pairs a browser-based Workspace with a desktop Launcher so agents running on your own machine or a remote server can join the same threads, files and tasks as your teammates.
Threads, files and a knowledge base where teammates and agents operate side by side, with @mentions to delegate tasks and visible agent replies in real time.
Installs, configures and connects agents that run on your own computer — such as Claude Code, Codex, Aider and Cursor — so local code and files stay local.
Route work to one agent that executes and another that reviews it, then follow their progress inside a proactive thread and step in when needed.
Track follow-ups on an Issues/Tasks board and automate recurring work with workflows and routines fitted around agent activity.

在你自己的電腦上執行一支 AI 代理團隊
每個代理都有自己的角色、指令與工作區,可以互相傳訊、交接任務、共用檔案並複核彼此的輸出。
讓 Codex、Claude Code、Gemini、Grok、OpenCode、Cursor 和 Cline 執行在你已付費的訂閱上,也可接入 ACP 代理、OpenAI 相容端點、Ollama 或 LM Studio。
把代理、模型與人工審核步驟組合成可重複使用的工作流,並可隨流程變化隨時調整。
工作區、對話、檔案與瀏覽器資料都保留在執行 OpenBot 的電腦上,而不是廠商的伺服器上。

Delivers a terminal-first coding agent optimized for local open-weight models, but lacks a graphical multi-agent management dashboard.
Delivers a terminal-first coding agent optimized for local open-weight models, but lacks a graphical multi-agent management dashboard.

免費 / 開源Executes model instructions directly inside local terminal sandboxes, but lacks a dedicated collaborative team management interface.
免費 / 開源Executes model instructions directly inside local terminal sandboxes, but lacks a dedicated collaborative team management interface.
免費 / 開源透過 Git 工作樹平行執行多個 CLI 編碼代理程式,但專注於基於終端機的編碼工作流程。

免費 / 開源統籌開源桌面 AI 工作團隊來處理瀏覽器與終端機任務,但缺乏預先設定的團隊協作範本。