OpenClaw: your open-source personal AI agent
A free, open-source AI agent that runs on your own computer and does real work from the chat apps you already have open: WhatsApp, Telegram, Discord, Slack, Signal, iMessage. Plug in Unifically and it can generate video, images and music too.
About the project
What is OpenClaw?
OpenClaw is a free, open-source AI agent that runs jobs through large language models and uses messaging apps as its main interface. Peter Steinberger and a growing community of contributors built it in TypeScript under the MIT license.
Instead of opening a browser tab, you message it in WhatsApp, Telegram, Discord, Slack, Signal or iMessage. From there it browses the web, runs shell commands, manages email, writes files, edits your calendar and calls any Skill you wire up, including Unifically's video, image and audio APIs.
A heartbeat fires every 30 to 60 minutes, so the agent can act between your messages: run scheduled jobs, work on long tasks in the background, and reach out when it needs you. People run it as a personal assistant, a company assistant, a family assistant and a round-the-clock dev teammate, all from one chat thread.
OpenClaw is an independent open-source project. Unifically isn't affiliated. We just like what they're building, and our APIs work well as one of the Skills.
License
MIT
Runs on
Your machine
Channels
10+
Source
TypeScript
Capabilities
What OpenClaw can do
A short tour of the main features, taken straight from the official OpenClaw site and docs.
Runs on your machine
Mac, Windows or Linux. Use Anthropic, OpenAI or a local model. Context, memory and Skills stay on your hardware.
Any chat app
Talk to it on WhatsApp, Telegram, Discord, Slack, Signal, iMessage, Teams or Matrix. DMs and group chats both work.
Persistent memory
Learns your preferences, your projects and your tone of voice over time. The memory survives restarts and new chat threads.
Browser & system access
Browses the web, fills forms, scrapes data. Reads and writes files, runs shell commands, executes scripts. You choose between full access or a sandbox.
Proactive heartbeats
A heartbeat fires every 30 to 60 minutes between your messages. The agent runs scheduled jobs, picks up background work, sends reminders and pings you when something needs attention.
Skills & plugins
Install community Skills or write your own. OpenClaw can even author them itself. Unifically ships as one of those Skills, giving the agent a media-generation backend.
Where it lives
Works in the chat apps you already use
Nothing new to install for chatting with the agent. OpenClaw connects to your existing messaging apps through built-in channels and community plugins.
Quick start
Install OpenClaw
Pick the install path that matches how you usually set up new tools. Each one takes a few minutes.
macOS and Linux. Installs Node.js for you, no setup needed.
curl -fsSL https://openclaw.ai/install.sh | bashWorks on macOS, Linux and Windows. Requires Node.js 22.14+ or Node 24. Switch channels later with openclaw update --channel dev or openclaw update --channel stable.
The Unifically Skill
Power your OpenClaw with affordable AI Models
OpenClaw routes prompts to LLMs but doesn't generate video, photoreal images, music or voice itself. Add Unifically as a Skill and your agent can produce all of it from any Telegram or Discord message.
Three steps to add Unifically to OpenClaw
You don't have to write the Skill by hand. OpenClaw will write it for you.
- 1
Grab a Unifically API key
Sign up and create a key from the API keys page. Pay per generation, no subscription, and up to 75% cheaper than calling the providers directly.
- 2
Ask your OpenClaw to build a Skill
In any chat, say something like "Build me a Skill that calls the Unifically API at api.unifically.com to generate video, images and audio. Here is my key: ..." and OpenClaw will write, test and hot-reload the Skill on its own.
- 3
Start generating from any message
"Make me a Veo clip of a dolphin surfing." "Generate a Suno track for my podcast intro." "Voice this paragraph in ElevenLabs." OpenClaw passes the call to Unifically, drops the result back in chat, and remembers what you like.
In the wild
What people are building
A few public OpenClaw use cases. Quotes are taken from posts on X with attribution to the original authors.
FAQ
Frequently asked questions
Short answers to the questions we get most often about OpenClaw and how Unifically fits in.
OpenClaw is a free, open-source AI agent that runs on your own computer. You message it through apps you already use (WhatsApp, Telegram, Discord, Slack, Signal, iMessage and others) and it does the work through large language models. Think of it as a personal Jarvis for your Mac, Linux box, Windows PC or Raspberry Pi.
Yes. OpenClaw is open source under the MIT license. You only pay for the LLM provider you pick (Anthropic, OpenAI, a local model, or a gateway) and any external APIs your Skills call, such as Unifically.
OpenClaw doesn’t lock you into one provider. People run it on Claude (Anthropic), GPT models (OpenAI), and a long list of local and proxy setups. Switching providers later won’t erase your memory or Skills.
No. OpenClaw has guided onboarding and a browser-based control UI. The one-line installer takes care of Node.js and the rest. If you can paste a command into a terminal, you can run it.
OpenClaw routes prompts to LLMs but it doesn’t generate video, photoreal images, music or voice on its own. Unifically exposes 80+ AI models behind a single HTTP API: Veo, Kling, Suno, GPT Image, Nano Banana, ElevenLabs and more. Add Unifically as a Skill in OpenClaw and the agent can produce media from any chat.
No. OpenClaw is an independent open-source project by Peter Steinberger and its contributors. We aren’t affiliated. We just think Unifically is the simplest way to give an OpenClaw agent video, image, music and voice generation.
OpenClaw itself runs entirely on hardware you control. Media generation still goes through Unifically (or another provider) over HTTPS, but your context, memory and Skills stay on your computer.
Install OpenClaw, then plug in Unifically.
Set up your own personal AI agent in minutes. Add Unifically right after for video, image and audio generation.