2026 年最值得自架的 OpenClaw 替代方案
七款開源的 OpenClaw 替代方案,依照大家離開它的原因排名:預設沒有沙箱、程式碼庫龐大、遙測,以及硬體需求。
概覽
| 機器人 | 授權 | 聊天 App | |||||
|---|---|---|---|---|---|---|---|
| OpenClaw | MIT | 39.4萬 | 是 | 否 | Telegram, WhatsApp, iMessage, Discord, Slack | 免費 | |
| ZeroClaw | MIT OR Apache-2.0 | 3.3萬 | 是 | 是 | Telegram, WhatsApp, Discord | 免費 | |
| NanoClaw | MIT | 3.1萬 | 是 | 是 | Telegram, WhatsApp, iMessage, Discord, Slack | 免費 | |
| Hermes Agent | MIT | 26.2萬 | 是 | 否 | Telegram, WhatsApp, iMessage, Discord, Slack | US$20/月起 | |
| Rakazo | Apache-2.0 | 7538 | 是 | 是 | Telegram, WhatsApp, iMessage, Slack | US$10/月起 | |
| NemoClaw | Apache-2.0 | 2.3萬 | 是 | 是 | Telegram, Discord, Slack | 免費 | |
| PicoClaw | MIT | 3萬 | 是 | 否 | Telegram, WhatsApp, Discord, Slack | 免費 | |
| nanobot | MIT | 5萬 | 是 | 否 | Telegram, WhatsApp, Discord, Slack | 免費 |
檢查清單
OpenClaw 是預設的開源代理:採用 MIT 授權,截至 2026 年 11 月 8 日有 394,193 顆 GitHub 星、29 個聊天頻道,每種桌面和手機平台都有 App。大多數尋找替代方案的人,並不是對它的能力不滿意,而是對它做事的方式感到不安。
大家為什麼離開 OpenClaw
常被提到的原因有四個,它們決定了下面的排名。
預設沒有沙箱。 OpenClaw 的沙箱開箱時是關閉的,所以除非你自己開啟核准機制或沙箱,否則代理會在你的主機上執行指令。再加上它能存取你的電子郵件、訊息和第三方技能,提示詞注入就成了真實的風險。它自己的維護者也警告,對非技術使用者來說它有風險。我們在 OpenClaw 安全嗎?中有更詳細的說明。
規模與複雜度。 142 個官方外掛以及 64 個模型和媒體供應商讓它功能強大,但這也代表有大量程式碼需要信任並持續更新,而且安裝過程預設你會用命令列。
遙測。 它預設每天連線到專案伺服器,傳送你的 OpenClaw 版本、作業系統、Node.js 版本和 CPU 架構,外加安裝計數和更新報告。影響不大,update.checkOnStart: false 就能關閉,但在你關閉之前它一直開著。
硬體與執行環境。 它需要 Node 24.16+ 或 26.1+,在筆電上沒問題,在 Raspberry Pi 或路由器上就沒那麼理想了。
我們依照這些替代方案能解決多少上述問題來排名,同時它們仍要做好核心工作:一個你從聊天應用程式與它對話、全天候運作的代理。
1. ZeroClaw:最接近「預設就安全的 OpenClaw」
ZeroClaw 解決了全部四個問題。它是單一 Rust 執行檔,只有一個 TOML 設定檔,能在 macOS、Linux、Windows、FreeBSD、Android 或 Raspberry Pi 上執行。它建議的 Balanced 設定檔會在中風險動作前先詢問、封鎖高風險動作、把檔案存取限制在工作區內,並在作業系統沙箱中執行工具(macOS 上是 Seatbelt,Linux 上是 Firejail)。它的文件聲明不傳送任何遙測。
問題在於頻道:Slack、iMessage 和 Signal 不在標準版本中,你要啟用額外功能從原始碼建置。如果你主要用 WhatsApp、Telegram 或 Discord,這就無關緊要。完整比較:ZeroClaw vs OpenClaw。

2. NanoClaw:真正的容器隔離,精簡的程式碼庫
NanoClaw 一開始是 OpenClaw 的一個更小、具備容器隔離的版本。每個代理都在自己的 Docker 容器中執行,只能看到你掛載的資料夾,對外的 API 呼叫會經過一個憑證閘道,所以代理永遠不會持有原始金鑰。主機端只有一個 Node 程序和少數幾個檔案。
它排在 ZeroClaw 之後有兩個原因:它以 Claude 和 Claude Code 為核心打造(客製化代表讓 Claude Code 修改你的 fork,而且沒有設定檔),以及除非你設定 NANOCLAW_NO_DIAGNOSTICS=1,否則它會傳送匿名的安裝診斷資料。請參考 NanoClaw vs OpenClaw。

3. Hermes Agent:轉換最輕鬆
Nous Research 的 Hermes Agent 在設計時就考慮到 OpenClaw 使用者。它能用一條指令匯入你的 OpenClaw 設定、記憶和技能,支援 20 多個聊天平台,共享指標預設關閉。它還增加了 OpenClaw 並未主打的能力:完成複雜任務後會自己撰寫並改進技能。
不過它沒有解決沙箱問題。終端機後端預設為本機,容器隔離是選用的。選它是為了遙測和功能,而不是安全性。更多內容請見 Hermes vs OpenClaw。

4. Rakazo:每個機器人一台沙箱電腦
Rakazo(由我們開發)為每個機器人提供一台自己的沙箱電腦,附瀏覽器、終端機和桌面,所以代理的指令在那裡執行,而不是在你的主機上。它採用 Apache-2.0 授權,自架安裝不傳送產品分析資料,並能透過 Ollama 或 LM Studio 執行本機模型。哪些動作需要你核准,由你設定。
它比上面幾個選項更吃資源(需要 Docker Engine 搭配 Compose),仍處於 beta 階段,而且 Telegram、WhatsApp 和 iMessage 只有在自架時才能使用。

5. NemoClaw:保留 OpenClaw,加上籠子
如果你喜歡 OpenClaw,只是不想讓它在你的機器上毫無約束地執行,NemoClaw 是 NVIDIA 用來在 OpenShell 沙箱中執行 OpenClaw 的技術堆疊。未經核准的網路目的地在操作者核准之前都會被封鎖,憑證也保存在沙箱之外。
它會讓另外三個問題變得更糟:至少 8 GB 記憶體(建議 16 GB)、需要容器執行環境、除非設定 NEMOCLAW_DISABLE_TELEMETRY=1 否則會傳送遙測給 NVIDIA,而且處於 alpha 階段。請參考 NemoClaw vs OpenClaw。

6. PicoClaw:適合極小的硬體
Sipeed 的 PicoClaw 是單一 Go 執行檔,至少需要 10 MB 可用記憶體(建議 32 MB 以上),能在 x86、ARM、RISC-V 和 MIPS 上執行,包括 10 美元的開發板、路由器和舊 Android 手機。文件中沒有記載任何遙測,支援大約 19 個聊天頻道。
它入選純粹是因為硬體。程序隔離預設關閉,專案本身也說在 v1.0 之前不要用於正式環境。

7. nanobot:小巧易讀的 Python
香港大學資料智慧實驗室的 nanobot 是一個精簡的 Python 代理,有精緻的瀏覽器介面、終端機介面、排程自動化和許多聊天頻道。它沒有遙測,支援 Ollama、vLLM 或 LM Studio。
和 OpenClaw 一樣,在你開啟 bubblewrap 或 Seatbelt 之前,它的 shell 指令都沒有沙箱,而 Windows 上完全沒有沙箱。如果你想要一個一個下午就能讀完的程式碼庫,就選它。

該選哪一個
- 想要 OpenClaw 的功能,同時有安全預設值且沒有遙測:ZeroClaw。
- 想讓每個代理都在容器中,而且不介意 Claude Code:NanoClaw。
- 想保留你的 OpenClaw 設定並停止遙測:Hermes Agent。
- 想繼續使用 OpenClaw 本身並放進沙箱,而且記憶體夠用:NemoClaw。
- 在開發板或舊手機上執行:PicoClaw。
想看一份不以 OpenClaw 為中心的更完整清單,請參考最佳自架 AI 代理。
替代方案
Personal AI agent in one Rust binary, MIT or Apache-2.0, with local or cloud models, 30+ channels, approval gates and an OS sandbox on by default.
Meets every check: one Rust binary, approval gates and an OS sandbox on by default, no telemetry, local models.
前往 ZeroClaw優點
- Safe defaults: the recommended profile asks before medium-risk actions, blocks high-risk ones, keeps files in the workspace and runs tools in an OS sandbox
- One small Rust binary with one config file, running on macOS, Linux, Windows and boards like a Raspberry Pi
- More than 30 channels, plus a web dashboard for chat, memory, config and cron
- Provider fallback chains, local models through Ollama, and cryptographic receipts for tool calls
- States plainly that it sends no telemetry and needs no cloud account or license server
缺點
- Slack, iMessage, Signal and several other channels are not in the standard release; you build from source with extra features to get them
- `zeroclaw quickstart` only sets up the model provider; channels and risk profiles mean editing the TOML config file
- The Docker sandbox backend isn't picked automatically under the default native runtime
- Model costs are on you
Small, MIT-licensed OpenClaw alternative that runs each agent in its own Docker container and chats with you on WhatsApp, Telegram, Slack, Discord and more.
Every agent in its own container and a tiny codebase, but tied to Claude Code and sends setup diagnostics by default.
前往 NanoClaw優點
- Every agent runs in its own Docker container and sees only the folders you mount, so shell access stays off your host
- Small codebase (one host process and a handful of files) that you customize by having Claude Code edit your fork
- Agents never hold raw API keys: outbound calls go through a credential gateway with per-agent policies and rate limits
- Many chat channels, including WhatsApp, Telegram, Slack, Discord, iMessage, Teams, Matrix and email, with one or several agents per channel
- Scheduled tasks with optional script gates that skip waking the agent when there's nothing to do
缺點
- Built around Claude and Claude Code: customizing, debugging and adding channels all go through Claude Code
- Sends anonymous setup diagnostics unless you set NANOCLAW_NO_DIAGNOSTICS=1
- No config files by design, so every change means editing code in your own fork
- Needs Docker, and Windows only works through WSL2
Free, MIT-licensed AI agent from Nous Research that runs on your own machine or server, learns reusable skills, and chats with you on Telegram, Slack and more.
MIT licensed, no telemetry and lives in your chat apps, but commands aren't sandboxed by default.
前往 Hermes Agent優點
- Free and MIT licensed, and runs on hardware you control
- Writes and refines its own skills from experience, on top of persistent memory and cross-session search
- One gateway for 20+ chat platforms, including Telegram, Discord, Slack, WhatsApp, Signal and email
- Run several separate agents as profiles, each with its own model, memory and skills, and let them talk in group chats
- Model-agnostic, with a one-command import from OpenClaw
缺點
- Setup expects comfort with the command line and with running your own server
- Model costs are on you unless you use free models, and hosted web search, images and speech need a paid Nous Portal plan or your own keys
- iMessage needs a separate BlueBubbles server on a Mac
- No native mobile app; Android support is a command-line install through Termux
Open-source, self-hostable platform for persistent AI bots that each get a job, a computer, memory and routines, using the model you choose.
Our own product. Meets every check: Apache-2.0, sandboxed bot computers, no telemetry by default, any model you connect.
前往 Rakazo優點
- Open source (Apache-2.0) and fully self-hostable, including with local models
- Choice of model per bot, including existing Claude or ChatGPT subscriptions
- Bots keep working on schedules and triggers with their own cloud computer
- Approval controls over which actions a bot may take on its own
缺點
- Still in beta, with frequent releases
- You supply and pay for the model on the hosted plans, unless you buy Rakazo credits
- Telegram, WhatsApp and iMessage/SMS are only available when self-hosted
- More setup than a single chat app
NVIDIA's Apache-2.0 stack for running OpenClaw, Hermes or Deep Agents inside locked-down OpenShell sandboxes with managed inference. Alpha.
Puts OpenClaw or Hermes in a locked-down sandbox, but it's alpha, heavy to run and sends telemetry by default.
前往 NemoClaw優點
- Puts OpenClaw or Hermes Agent in a sandbox with network, filesystem and process controls instead of running it on your host
- Unapproved network destinations are blocked and surfaced for an operator to approve
- Inference, messaging and MCP credentials stay outside the sandbox
- Backed by NVIDIA, with local inference through Ollama, vLLM or NIM and a tested path on DGX Spark
缺點
- Alpha: maintainers support it on a best-effort basis, and several channels are experimental
- A wrapper, not an agent: what the bot can do depends on the OpenClaw or Hermes version inside
- Heavier than running the agent alone: 8 GB RAM minimum, a 2.4 GB sandbox image and a container runtime
- The CLI sends operation telemetry to NVIDIA unless you set NEMOCLAW_DISABLE_TELEMETRY=1
Free, MIT-licensed personal AI agent written in Go that runs as one small binary on cheap boards, old phones and PCs, and chats on Telegram, Discord and more.
Runs on cheap boards in a few dozen MB of RAM with no telemetry, but it's pre-1.0 and commands aren't sandboxed by default.
前往 PicoClaw優點
- Tiny footprint: one Go binary that Sipeed says boots in under a second and needs only 10 to 20 MB of RAM
- Runs on hardware others can't, from $10 RISC-V boards and routers to old Android phones
- About 19 chat channels, including Telegram, Discord, Slack, WhatsApp, Matrix, LINE and Delta Chat
- Web search works without an API key through a built-in DuckDuckGo fallback
- Heartbeat check-ins, cron reminders, MCP tools and routing between several agents
缺點
- The project warns it is in early rapid development, may have unresolved security issues, and should not be used in production before v1.0
- The default workspace restriction only checks paths; true process isolation is off by default and implemented only on Linux and Windows
- Native WhatsApp needs a custom build or a separate bridge
- No built-in image generation, and voice messages are only transcribed if you add a provider such as Groq
Free, MIT-licensed personal AI agent in Python with a browser UI and terminal app, scheduled tasks, memory, and chat on Telegram, Discord, Slack and more.
Small, readable Python with a good web UI and no telemetry, but shell commands aren't sandboxed by default.
前往 nanobot優點
- Free and MIT licensed, with a small Python core that is meant to be read and modified
- Polished bundled WebUI with projects, file diffs, an MCP app catalog and an automations calendar, plus a terminal UI
- Many chat channels, including Telegram, Discord, Slack, WhatsApp, WeChat, Feishu, Teams, Matrix and email
- Scheduled automations, long-running goals and a heartbeat that reports only useful results to your last chat
- Works with local models through Ollama, vLLM or LM Studio
缺點
- Shell commands are not sandboxed by default; the bubblewrap or macOS Seatbelt sandbox must be turned on, and Windows has none
- Model costs are on you, and voice messages need a separate transcription provider such as Groq or OpenAI
- No native mobile or desktop app; on a phone you use the WebUI in a browser or a chat app
- Missed scheduled runs are not replayed and interrupted subagent tasks are not relaunched after a restart
功能
來源
OpenClaw
- OpenClaw homepage and Foundation section
- OpenClaw integrations: channels, providers and plugins
- openclaw/openclaw on GitHub (MIT, v2026.11.1)
- OpenClaw docs: Sandboxing (off by default)
- Wikipedia: OpenClaw
- Notebookcheck: OpenClaw can burn through hundreds of dollars per day
- OpenClaw docs: Memory overview
- OpenClaw docs: Automations
- OpenClaw docs: Chat channels
- OpenClaw docs: iMessage
- OpenClaw docs: Heartbeat, now a default Proactive check job
- OpenClaw docs: Multi-agent routing
- OpenClaw docs: Platforms (macOS, Windows, Linux, iOS and Android apps)
- OpenClaw docs: Tools (exec, browser, web search)
- OpenClaw docs: Audio and voice notes (inbound transcription)
- OpenClaw docs: Image and media support (inbound media)
- OpenClaw docs: Install (system requirements, installer script, npm, desktop apps, Docker)
- OpenClaw docs: One-click deploy on Railway
- OpenClaw docs: Usage telemetry and update checks (on by default, feature statistics opt-in)
- OpenClaw docs: Ollama provider (local models)
ZeroClaw
- zeroclaw-labs/zeroclaw on GitHub (MIT OR Apache-2.0, v0.8.7, README)
- ZeroClaw website
- ZeroClaw docs: Sandboxing
- ZeroClaw source: risk presets (Balanced, recommended, has the sandbox on)
- ZeroClaw docs: Channels overview (default and standard release bundles)
- ZeroClaw docs: You own it (no telemetry)
- ZeroClaw docs: Docker and containers
- ZeroClaw source: tools (image generation, screenshot, web)
NanoClaw
Hermes Agent
- NousResearch/hermes-agent on GitHub (MIT, v0.23.0)
- Hermes Agent homepage and Nous Portal pricing
- Hermes Agent documentation
- Hermes Agent docs: Profiles, running multiple agents
- Hermes Agent docs: Bot Mode
- Hermes Agent docs: Voice mode
- Hermes Agent docs: Vision and image paste
- Hermes Agent docs: iMessage via BlueBubbles
- Nous Research on X: Meet Hermes Agent (February 25, 2026)
- Hermes Agent docs: Session heartbeats and cron watchdogs
- Hermes Agent docs: FAQ (Is my data sent anywhere?)
- Hermes Agent docs: Installation (desktop packages, install script, Docker)
- Hermes Agent docs: Platform support (no Linux desktop package; PyPI installs unsupported)
- Hermes Agent docs: Docker setup
- Hermes Agent docs: Configuration (terminal backend defaults to local)
- Hermes Agent docs: Security (approval modes, optional container isolation)
- Hermes Agent docs: Local models
- Hermes Agent example config (shared metrics off by default)
Rakazo
- Rakazo homepage
- Rakazo pricing
- Rakazo privacy policy
- Rakazo on GitHub (README and license)
- Rakazo changelog
- Rakazo releases
- Rakazo docs: Image and video generation
- Rakazo docs: Memory
- Rakazo docs: Connect apps
- Rakazo docs: Routines
- Rakazo docs: Chat apps
- Rakazo docs: Messaging apps (self-hosting)
- Rakazo docs: Bots and chats (add another bot, bot templates)
- Rakazo docs: Groups (several bots in one chat)
- Rakazo docs: Mobile app (notifications)
- Rakazo privacy policy (updated Sep 14, 2026)
- Rakazo privacy policy source on GitHub
- Rakazo docs: Logs and monitoring (product analytics off by default)
- Rakazo docs: Desktop app (run locally in Docker, updates)
- Rakazo docs: Docker computers
- Rakazo docs: Self-hosted models (Ollama, LM Studio)
- Rakazo Railway template setup guide
NemoClaw
- NVIDIA/NemoClaw on GitHub (Apache-2.0, README: supported agents, alpha status)
- NemoClaw docs: Overview
- NemoClaw docs: Prerequisites and platforms
- NemoClaw docs: Choose an inference provider (Ollama, vLLM, NIM)
- NemoClaw docs index: messaging channels (WhatsApp, Teams, Google Chat, WeChat experimental)
- NemoClaw source: telemetry delivery and NEMOCLAW_DISABLE_TELEMETRY
PicoClaw
- sipeed/picoclaw on GitHub (MIT, README with security notice and channel list)
- PicoClaw releases (v0.0.1 on February 9, 2026; v0.3.1 on July 3, 2026)
- PicoClaw official website and downloads
- PicoClaw docs: Configuration (workspace restriction, heartbeat, voice, multi-agent routing)
- PicoClaw source: subprocess isolation, disabled by default
- PicoClaw docs: Tools configuration (web search, exec, cron, MCP, skills)
- PicoClaw docs: Chat apps (WhatsApp native build or bridge)
- PicoClaw docs: Hardware compatibility and minimum requirements
- PicoClaw docs: Docker
nanobot
- HKUDS/nanobot on GitHub (MIT, v0.4.1, README)
- nanobot documentation
- nanobot release archive (officially launched February 2, 2026)
- nanobot docs: Chat apps
- nanobot docs: Providers and models (Ollama, vLLM, LM Studio)
- nanobot docs: Configuration (security, exec sandbox off by default, transcription, Langfuse)
- nanobot docs: Automations and heartbeat
- nanobot docs: Concepts (heartbeat delivery, subagent recovery)
- nanobot docs: Memory and Dream
- nanobot docs: Multiple instances
- nanobot docs: Deployment (Docker, services, Render)
- nanobot source: skills installer runs with DISABLE_TELEMETRY







