Related Topics
ASUS Zenni Claw User Installation & User Guide
ASUS Zenni Claw is an AI agent platform built to make Agentic AI simple, approachable, and trustworthy. Instead of just answering questions one by one like a chatbot, it helps you effortlessly breeze through complete, multi-step tasks across your work, daily life, travel, and creative projects.
A standard chatbot simply responds to a prompt. ASUS Zenni Claw is built to deliver a finished outcome. It actually understands your ultimate goal, maps out the necessary steps, triggers specific skills, and handles complex, multi-step workflows for you from start to finish.
3. Set Up AI Providers
4. Set Up Channels
5. Start to Chat
6. Pre-Built ASUS Tasks & Skills
7. Settings
Before installing ASUS Zenni Claw, the installer automatically checks whether your device can run the AI environment (WSL2, Ubuntu, a local LLM, and Docker). No terminal or manual configuration is required.
Minimum requirements:
- OS: Windows 11
- RAM: 16 GB or more
- Storage: 20 GB or more (40 GB+ recommended when using local AI)
Local AI requirements(optional):
- Option 1 — Discrete GPU: NVIDIA GeForce RTX 5090 with 24 GB+ vRAM, 16 GB+ RAM, and 40 GB+ storage.
- Option 2 — Integrated AI processor: Intel Core Ultra X9 388H or AMD Ryzen AI Max+ 300 series, with 32 GB+ RAM and 40 GB+ storage.
2.1 Auto-configure Al environment
(1) Launch the ASUS Zenni Claw installer. On the Welcome screen, the app runs a ①[System Check] and reports your Operating System, RAM, and Storage.
(2) When every item passes, you will see ②[Your system meets the installation requirements]. Click ③[Continue] to proceed.
(3) If an item fails (for example, not enough storage), the installer shows ④[Your device is incompatible. Cannot proceed with installation]. Free up resources and run the check again.
(1) Next, the installer ⑤[checks the Local AI Provider hardware] — GPU, vRAM, RAM, and Storage — to confirm your device can run a local model.
(2) When all requirements are met, you can tick ⑥[Would you like to install a local LLM?] to install the local AI. Click ⑦[Continue] to proceed.
(3) If an item fails, the installer shows ⑧[Incompatible device. Local Al unavailable. Other settings remain accessible].You can still use the app with cloud models.
(1) The installer ⑨[runs each steps automatically]: enabling WSL2, installing Ubuntu 22.04 LTS, the local AI provider, Guardrail (password / injection protection), LiteLLM, and Skills. A progress bar shows the overall status.
(2) Some steps require a system restart. When prompted, save your work and click ⑩[Restart Now (or Restart Later)]. Installation resumes automatically after the reboot.
(3) When every component passes, the screen shows “ASUS Zenni Claw is Ready,” with your Local LLM and Guardrail status. Click ⑪[Start] to open the app.
(1) When opening the app, the app will require the user to agree to the Privacy Policy & AI Terms & Conditions.
(2) The app also requires the user to set up an AI Provider and channel; the user can skip this for now and set them up later.
ASUS Zenni Claw works with both cloud AI providers and on-device (local) models. You manage them on the Models page, where you can also monitor token usage and configure Auto-Routing between cloud and local engines.
(1) Open ①[Models] from the left navigation. The page lists your connected AI Providers, the Auto-Routing rule, and Token Usage. Click ②[+ Add Provider] to add a new one.
(2) ASUS Zenni Claw works with both cloud AI providers and on-device (local) models. You manage them on the Models page, where you can also monitor token usage and configure Auto-Routing between cloud and local engines.
(3) To add a cloud provider, select the ③[Cloud LLM tab]. Choose a provider (Anthropic, OpenAI, or Gemini), pick a ④[model], then enter your ⑤[API key].
◉For the method to obtain an API Key, please refer to the related FAQ: How to get API Keys
(4) Click ⑥[Verify]. When the key is valid you will see a confirmation such as ⑦[Verified — claude-sonnet-4-6 is available] Click ⑧[Continue] to save.
(5) To use a local model, open the Local LLM tab. Once installed, when all requirements are met, you can tick ⑨[Would you like to install a local LLM?] to install the local AI. Click ⑩[Continue] to proceed.
(6) Use Auto-Routing to set a default Cloud LLM and a Local LLM so each task runs on the most suitable engine. Click ⑪[Edit], choose your ⑫[models], then click ⑬[OK].
Channels let you talk to ASUS Zenni Claw through third-party messaging apps — Telegram and WhatsApp. You add and manage them on the Channels page.
(1) Open ①[Channels] from the left navigation. Configured channels show their status (Connected / Disconnected). Under Supported Channels, click ②[Connect] for the service you want to add.
(2) Telegram: create a bot with @BotFather, copy the ③[Bot Token], get your ③[Chat ID] from @userinfobot, then paste both into the app. Use ④[+] to register more than one user ID.
◉For the method to use it, please refer to the related FAQ: How to Set Up Channels
(3) WhatsApp: Click the ⑤[Generate QR Code]. Open WhatsApp on your phone, go to Settings > Linked Devices > Link a Device, and scan the ⑥[QR code] shown in the app. A “Success!” message confirms the link.
◉For the method to use it, please refer to the related FAQ: How to Set Up Channels
(4) To control who can use a bot, click ⑦[Edit] on the channel and manage the user list — add or remove ⑧[Chat IDs] as needed, then ⑨[Save].
The Chat page is where you interact with the agent. Choose a session and model, type a request, attach files, or start from a ready-made prompt.
(1) Open ①[Chat]. Choose your ②[Session & Model] at the top, then type a request in the message box, or pick a suggested prompt under ③[Work / Life / Travel Assistant]. Click ④[Generate (or press Enter)].
(2) The agent replies in the conversation view. Tool calls ⑤[appear inline]; expand an entry to see what the agent did.
(3) Click a tool entry to view its ⑥[detailed input and output], including status and any returned data.
(4) Add context by ⑦[dragging files] into the conversation, or use the attach button. The app shows the files queued for your next message.
(5) Press the search ⑧[/command icon] to open the command menu, where you can run quick commands such as /status, /restart, or /clear.
6. Pre-Built ASUS Tasks & Skills
ASUS Zenni Claw ships with ready-made prompts grouped into Work, Life, and Travel assistants, plus a library of Skills the agent can call to get things done.
(1) Open the ①[Prompt Library] to browse ready-made tasks — for example Slide Generator, Industry News Digest, Meeting Actions, Morning Briefing, Outdoor Gear Advisor, Flight Monitor, and Trip Itinerary Planner. Select one to start.
(2) Go to ②[Skills] to browse and manage the agent’s capabilities. Use the ③[search box] to find a specific one; use the ④[toggle button] on each Skill to enable or disable it.
(3) Skills are organized into ⑤[Workspace Skills] and ⑥[Built-In Skills]. Each Skill shows a ⑦[short description and the channels or tasks] it supports.
(1) Open ①[Settings] and click on ②[General]. Activate the ③[toggle button] to enable it, and the Zenni Claw will automatically start upon boot to provide continuous background service.
(2) Click ④[Launch] to open the ASUS Zenni Claw Update Center and ⑤[review available updates]. Critical Updates are delivered as a release bundle; click ⑥[Update now] to apply them. Then, click ⑦[Check update].
(1) Click the ①[User Guide] and view the contents of FAQs
(2) Click the ③[About] and check ④[SW Version]. It tells you whether the app is up to date and when it was last checked.
