OpenClaw (aka Moltbot/Clawdbot)
Peter Steinberger / OpenClaw Foundation · Other
OpenClaw (also known as Moltbot or Clawdbot) is an open-source AI assistant that runs on your own computer and meets you inside the chat apps you already use, such as Discord, Slack, Telegram, and WhatsApp. Instead of living in a browser tab, it keeps your state, memory, and credentials on your own hardware and lets you swap the underlying model whenever you want. It's a self-hosted AI agent for people who want an assistant that actually gets things done without handing their workflows to a hosted service. Sound appealing? Here's what you're signing up for.
About OpenClaw (aka Moltbot/Clawdbot)
What Is OpenClaw (aka Moltbot/Clawdbot)
OpenClaw is a local-first personal AI assistant, not a cloud chatbot. You install it on a Mac, Linux box, Windows machine, or a small VPS, and a component called the Gateway runs quietly in the background as the control plane for sessions, tools, events, and your channel connections. Then you talk to it in the apps you already have open instead of learning a new interface. That part feels natural fast. The setup isn't.
The project has a naming history that trips people up. It started as Clawdbot, briefly became Moltbot after a trademark dispute in January 2026, and settled on OpenClaw by the end of that month. All three names point to the same thing, and old commands still work, so don't worry if you find a tutorial using an older name.
Two things define the product. First, it's model-agnostic: Anthropic, OpenAI, MiniMax, OpenRouter, and local models are pluggable, so you're not locked to one provider. Second, it's open source under a permissive license and stewarded by a nonprofit foundation, with no paid tier or hosted service. The catch is that setup asks more of you than a typical app. You'll need Node.js comfort, a model API key, and a bit of patience the first time you wire up a channel. Not a huge deal for developers. A real hurdle for everyone else.
Getting Started
- Run the installer for your platform. On macOS, Linux, or WSL2 use the shell script, and on Windows use the PowerShell version, or install the published npm package if you already manage Node.js.
- Complete the onboarding wizard, which verifies model access, creates your workspace, and configures the Gateway.
- Connect at least one messaging channel, such as Telegram, Discord, or Slack, so you can reach the assistant from a chat app you use daily.
- Open the Control UI dashboard and send a test message to confirm the self-hosted AI agent is running and replying.
- Add skills or plugins as you go, and swap or add model providers whenever your needs change.
Product Information
A quick look at OpenClaw (aka Moltbot/Clawdbot)'s pricing, supported platforms, and performance.
Best for
The users, tasks, and scenarios where this tool fits best.
Users
- Privacy-minded individuals
- Developers comfortable with Node.js
- Small teams that trust each other
- Anyone who lives in WhatsApp
- Tinkerers who like to customize
Tasks
- Answering questions inside chat apps
- Organizing files and folders
- Summarizing and following up on email
- Running repeated scheduled jobs
- Prototyping builds and drafts
Scenarios
- Keeping a persistent assistant on a home server or VPS
- Working across several chat platforms at once
- Keeping data off the cloud entirely
- Automating a recurring personal workflow
Key features
Local-first architecture
OpenClaw stores state, memory, and credentials on your own machine, and the Gateway acts as the always-on control plane for sessions, tools, and connections. That design keeps your personal AI assistant self-hosted and gives you a concrete privacy boundary, since your prompts only reach the model providers and chat platforms you choose to configure.
Chat-app connectivity
Rather than forcing a new interface, OpenClaw meets you where you already talk. It supports Discord, iMessage, Slack, Teams, Telegram, WhatsApp, Google Chat, Signal, Mattermost, and more than twenty other channels, plus native apps for macOS, iOS, Android, Windows, and Linux. You message it like a contact, and it replies in the same thread. No new app to learn.
Model-agnostic provider support
You're not tied to one AI vendor. Anthropic, OpenAI, MiniMax, OpenRouter, and local models are all options, with per-agent routing and failover so a backup model can take over if your primary one is unavailable. Swapping providers doesn't require rebuilding the rest of your setup. That flexibility matters more than it sounds.
Skills and plugins
OpenClaw extends through skills, which package reusable capabilities like browser access, file handling, and code generation. You can install community skills or write your own, which is how the assistant grows from answering questions into executing multi-step tasks on your behalf.
Multi-agent routing
One OpenClaw install can run several agents, each tied to a different channel, account, or task with its own workspace and defaults. That lets you keep a personal agent separate from a work one while sharing the same underlying Gateway.
Shared team deployment
The same Gateway that serves one person can serve a small team. Sessions track who created and steered them, the Control UI shows who's active, and configurable operator roles limit what each teammate can do. It's genuinely a first-class setup, not a hack.
Pros and cons
Pros
- Open source with no paid tier, hosted service, or token. You only pay for whatever model API you use.
- Runs on your own hardware. Memory, credentials, and session history stay local.
- Model-agnostic design means you can switch or add providers without rebuilding anything.
- Reaches you in chat apps you already use, including Discord, Slack, Telegram, and WhatsApp.
- Scales from a single personal assistant to a shared team workspace on the same Gateway.
Cons
- Setup is more involved than a hosted chatbot, since it needs Node.js, a model API key, and channel configuration.
- The privacy promise depends on your own choices, because prompts still go to whatever model provider you wire up.
- A shared Gateway is one trust domain, so it only works for teams that already trust each other.
- Newer features and channels move quickly, which means occasional rough edges and shifting documentation.
Frequently asked questions
Yes. There's no paid tier, hosted service, or token, and the project is open source under a permissive license. Your only real cost is whatever your chosen model provider charges for API usage, or nothing if you run local models.
Related content
Explore related tools, skills, and articles for OpenClaw (aka Moltbot/Clawdbot).
OpenClaw (aka Moltbot/Clawdbot) Alternatives
BinkBink
BinkBink · OtherBinkBink is a free online game platform and AI game maker that lets anyone turn a short text description into a playable browser game. You can jump into hundreds of community-made games. Or describe your own idea and play it in seconds, then share it with friends. Want to create your own game? You don't need to code. No engine setup, no download, no hassle.
Audiogen
Audiogen Inc. · OtherAudiogen is an AI music generator built by Audiogen Inc., a small research team that spent about 2.5 years training its own generative music model and designing a web interface around it. Instead of a plain text box, this AI music tool turns the timeline into a beginner-friendly Generative Audio Workstation, or GAW, where inpainting, extending, remixing and stem editing work more like painting on a canvas. The product is still in beta, so access runs through a waitlist or an invite. Paid plans aren't published yet.
Aiml API
AIMLAPI OÜ · OtherAiml API is a unified AI model API that puts more than 1000 models from OpenAI, Google, Anthropic, and others behind one endpoint and one bill. You write code against a single OpenAI-compatible schema, then switch between chat, image, video, and audio models by changing a model string. It suits developers and small teams who want multi-model access without juggling a dozen separate provider accounts, and it removes the usual billing headache that comes with testing several vendors. One key covers it all.