值得信赖的 Rust 原生 claw

One binary — sandboxed, built in Rust, yours to own.

Not a chatbot wrapper. Not a cloud service tied to one provider. A real agent server that runs on your hardware, remembers across sessions, and reaches you through every channel you use.

Get Started Documentation

Onboarding — secure your instance with passkey or password

Onboarding — setup summary with identity, LLM and channel config

语音 settings — speech-to-text and text-to-speech providers

Sandboxed shell execution in Docker container

Telegram bot channel with voice and events

Web UI showing Telegram channel messages

Up and running in 60 seconds

Three steps. No dependencies.

安装

$ curl -fsSL moltis.org/install.sh | sh

Configure

Open localhost:13131, paste your provider key, pick a model. OAuth providers like GitHub Copilot work with zero config.

Chat from anywhere

Web UI

Discord

Slack

Matrix

Nostr

Teams

Secure by design

Your keys and private data never leave your machine. Sandboxed by default — your AI can't touch your filesystem unless you say so.

Your hardware

Runs on a Mac Mini, a Raspberry Pi, or any server you own. Always on, always local — like a house elf for your digital life.

macOS

Linux

Windows

Full-featured

Telegram, WhatsApp, Discord, Slack, Matrix, Nostr, Teams, voice, memory, scheduling, tools — all built-in. No plugin marketplace to get supply-chain attacked through.

What your agent actually does

Not a demo. These run today.

Schedules tasks

Recurring cron jobs in natural language

Runs code safely

Sandboxed shell in Docker or Apple Containers

Browses the web

Fetches, summarizes, and SSRF-protected

Manages your calendar

CalDAV integration for events and reminders

Messages everywhere

Telegram, WhatsApp, Discord, Slack, Matrix, Nostr, Teams

Remembers context

Persistent memory with vector + full-text search

Talks and listens

语音 input and output with cloud and local providers

Extends itself

Creates its own skills, hooks, and MCP 工具 at runtime

Moltis worked in 1 command. I found several AI guardrails missing or turned off in OpenClaw which are working out of the box here. Moltis is way more performant and stable.

Noita on Discord #1 #2

I've found the project incredibly useful. In fact I do my Moltis contributions in Moltis.

Cstewart-HC on GitHub discussion

兼容你的技术栈

20+ 提供商、本地模型和所有主要渠道 — 全部内置。

云提供商

Anthropic

OpenAI

Google Gemini

Mistral

DeepSeek

Fireworks

Cerebras

MiniMax

Moonshot

Z.AI

Venice

NEAR AI Cloud

本地 & 自带模型

Ollama

LM Studio Local GGUF CPU/GPU

MLX Apple Silicon

Hugging Face

OpenRouter

Any OpenAI-compatible

OAuth — 零配置

GitHub Copilot

OpenAI Codex

Kimi Code

频道

Web UI

Discord

Slack

Matrix

Nostr

Microsoft Teams

iOS Soon

GraphQL API

JSON-RPC

Get Started

Install Moltis and start your AI assistant in minutes

~ quickstart

$ curl -fsSL https://www.moltis.org/install.sh | sh

Supported Platforms

Pre-built releases

macOS

Apple Silicon & Intel

Linux

x86_64 & ARM64

Windows

x86_64

Install with one command, or pick a release artifact for your platform from GitHub Releases.

Try it in the cloud

Ready in under a minute

Cloud is a great way to try Moltis. For the full experience, run it on your Mac, a VM, or any dedicated host.

Deploy on DigitalOcean

Fly.io

Need setup details? See Cloud Deploy docs.

功能

您需要的一切 in one binary

Single Binary

One self-contained binary. No runtime dependencies, just download and run.

Local LLMs

Run your own models locally. Automatic download and setup included.

Authentication

HTTPS by default. Password, token, and passkeys access.

Sandboxed Browsing

Run browser sessions in isolated Docker containers for safer automation.

SSRF Protected

Blocks loopback, private, and link-local IPs from LLM fetch.

Streaming-first

First token appears instantly. Smooth replies, even during long runs.

Skills, 钩子 & MCP

Plugins, hooks, MCP tool servers with OAuth 2.1. Stdio or HTTP/SSE, auto-restart.

FS or 沙盒

Full filesystem or per-session Docker, Podman, Apple Container, or WASM isolation.

Long-term Memory

Hybrid vector + full-text search. Your agent remembers context.

Pi-Inspired Self-Extension

Pi-inspired self-extension: creates its own skills at runtime. Session branching, hot-reload.

Multi-Channel

Web UI, Telegram, WhatsApp, Discord, Slack, Matrix, Nostr, Teams, or API. One agent, every frontend.

语音

Talk to your assistant with multiple cloud and local TTS/STT providers.

OpenClaw Import

新

Full read-only import of your OpenClaw workspace — conversations, skills, and settings. Switch in one click.

Multi-Agent Personas

Create multiple agent personas with distinct personalities and capabilities. Full CRUD management UI.

Encryption Vault

Encryption-at-rest vault for API keys and secrets. Your credentials are encrypted on disk, not just in memory.

iOS Companion

Soon

Native iOS app with push notifications via APNS relay. Coming soon to the App Store.

GraphQL API

Full GraphQL API exposing all RPC methods. Build your own integrations and frontends.

Internationalization

完整 i18n 支持。现已内置 11 种语言。

Webhooks

新

Run AI agents on GitHub PRs, Stripe payments, or any webhook. Review code, triage issues, handle payments — automatically.

And much more...

安全

✓ Passkeys (WebAuthn)
✓ Scoped API keys
✓ Encryption-at-rest vault
✓ Secrets zeroed on drop
✓ Human-in-the-loop approval
✓ Origin validation (CSWSH)
✓ No unsafe code

LLM Providers

✓ Anthropic, OpenAI, Gemini
✓ DeepSeek, Mistral, Groq, xAI
✓ OpenRouter, Ollama, Local LLM
✓ Custom OpenAI-compatible endpoints
✓ Provider fallback chains
✓ Per-provider metrics

Memory

✓ Hybrid search (vector + FTS)
✓ Local embeddings (GGUF)
✓ OpenAI batch API (50% off)
✓ Embedding cache
✓ File watching / live sync
✓ Session export

Extensibility

✓ MCP server support
✓ GraphQL API
✓ CalDAV calendar integration
✓ Hook system (lifecycle)
✓ Cron job scheduling
✓ Webhook triggers
✓ i18n（11 种语言）
✓ OpenClaw 导入
✓ TOML configuration

频道

✓ Web UI
✓ Telegram bot
✓ WhatsApp
✓ Discord
✓ Slack
✓ Matrix
✓ Nostr
✓ Microsoft Teams
○ iOS companion app (coming soon)
✓ JSON-RPC + GraphQL API
✓ Push notifications
✓ Multi-device sync

Observability

✓ Prometheus metrics
✓ OpenTelemetry tracing
✓ Structured logging
✓ Per-provider charts
✓ SQLite persistence
✓ Real-time WebSocket

安全优先设计

您可以信赖的机器 AI 助手

为什么您应该信任它？

Moltis 是一个文件。您下载它，运行它，完成。没有插件商店，没有包管理器，没有可能被破坏的运行时。整个应用程序是一个您可以验证的单一二进制文件。

您的 AI 默认在沙盒中运行。除非您明确允许，否则它无法读取您的文件、运行命令或浏览网页。每个工具操作都可以要求您的批准 — 或者您可以预先批准您信任的特定操作。

您的密码和 API 密钥被封装，永远不会被意外记录，并且在不再需要时内存会被清零。它所使用的编程语言（Rust）在编译器层面消除了整类安全漏洞。

这里您找不到的

没有可能被攻破的插件市场。没有 JavaScript 运行时。没有 1,200 个您无法审查的包的依赖树。一个您可以从头到尾审计的二进制文件。

我们如何保护您

如果…	Moltis 如何处理
有人发现您的实例	首次启动即需要密码或通行密钥，自动 HTTPS，速率限制
AI 尝试运行危险代码	一切在沙盒中运行（Docker 或 Apple Container）。每个操作都需要您先批准
依赖被攻破	没有可被攻破的 npm/pip 包。一个二进制文件，可审计的依赖。Telegram 等集成是内置的、经过测试的代码
AI 访问您的内部网络	内部和私有网络地址被自动阻止
有人试图窃取您的密钥	API 密钥和密码在日志中被隐藏，使用后在内存中被清零
会话被劫持	来源检查防止跨站攻击，默认启用 HTTPS
未经授权的访问	通行密钥、范围化 API 密钥、逐工具权限策略

沙盒执行

Docker & Podman

每个对话都有自己的隔离容器。Shell 命令和网页浏览在沙盒内进行，而不是在您的真实系统上。

Apple Containers

在 Mac 上，使用 Apple 的内置虚拟化实现轻量隔离 — 无需 Docker。

WASM 沙盒

基于 Wasmtime 的沙盒，具有通用故障转移。无需容器的轻量隔离。

完全开源

小

小型隔离 unsafe 表面*

npm 依赖

Rust 工作区 crate

~270K

Rust 代码行

470+

带测试的 Rust 文件

MIT

许可证

*Unsafe 代码隔离在 Swift FFI、本地模型包装器和预编译 WASM/运行时边界中。核心代理和网关路径保持安全 Rust。

Moltis 对比

您可以自托管的开源 AI 助手

一览

	OpenClaw	Hermes Agent	Moltis
语言	TypeScript + native apps	Python + TypeScript UI	Rust
运行时	Node.js + npm/pnpm/bun	Python + uv/pip	Single binary
Main shape	Gateway + apps + nodes	CLI + gateway + research tools	Persistent agent server
沙盒	App-level permissions	Local, Docker, SSH, cloud backends	Docker/Podman + Apple + WASM
认证	Pairing + gateway controls	CLI + gateway setup	Password + Passkey + API keys
语音	语音 wake + talk modes	语音 memo transcription	Built-in (15+ providers)
MCP	Plugin/integration support	MCP integration	Yes (stdio + HTTP/SSE)
记忆 / RAG	Plugin + context engine	Memory + session search + user model	SQLite + FTS + vector
钩子	Plugin/runtime hooks	钩子/plugins	15 event types
频道	Broad channel and device list	Telegram, Discord, Slack, WhatsApp, Signal, CLI	Web, Telegram, WhatsApp, Discord, Slack, Matrix, Nostr, Teams
API	Gateway APIs	CLI/gateway APIs	JSON-RPC + GraphQL
OpenClaw 导入	—	Migration command	Full (read-only, safe)

安全对比

	OpenClaw	Hermes Agent	Moltis
代码沙盒	App-level permissions and tool controls	Local, container, SSH, and cloud backends	Docker/Podman + Apple + WASM
密钥处理	Config, env, and plugin paths	Config and provider credentials	加密保险库，已编辑，已清零
SSRF 防护	Tool/plugin dependent	Tool/backend dependent	Blocks loopback/private/CGNAT
WebSocket 来源	Gateway dependent	Gateway dependent	跨域拒绝
钩子控制	Plugin/runtime hooks	钩子/plugins	检查 / 修改 / 阻止
速率限制	Gateway dependent	Gateway dependent	每 IP 节流 + 登录限制

代码库快照

	OpenClaw	Hermes Agent	Moltis
本地源码规模*	~1.1M app LoC	~152K app LoC	~270K Rust LoC
包形态	npm packages, extensions, apps	Python package, plugins, tools, UI packages	59 Rust 工作区 crate
主要安装方式	npm install -g openclaw	install script, then hermes	Install script, Homebrew, Docker, Cargo
测试覆盖信号	Large TS/app test tree	Python and TUI tests	470+ 带测试的 Rust 文件

*使用 tokei 从本地 checkout 统计，排除依赖目录和生成的构建输出。LoC 是规模和可审计性的信号，不是排名。

基于本地 checkout 快照和公开项目信息。项目发展迅速 — 请查看各仓库获取最新信息。文档中的完整对比.

值得信赖的 Rust 原生 claw

Up and running in 60 seconds

安装

Configure

Chat from anywhere

Secure by design

Your hardware

Full-featured

What your agent actually does

兼容你的技术栈

Get Started

Supported Platforms

macOS

Linux

Windows

Try it in the cloud

功能

And much more...

安全

LLM Providers

Memory

Extensibility

频道

Observability

安全优先设计

为什么您应该信任它？

我们如何保护您

沙盒执行

Docker & Podman

Apple Containers

WASM 沙盒

完全开源

Moltis 对比

一览

安全对比

代码库快照

What users are saying