Editorial illustration for Oppo open-sources X‑OmniClaw agent that scrolls, captures, reads prices in‑app
Oppo open-sources X‑OmniClaw agent that scrolls,...
Your phone is about to get pushy. Oppo has released the code for an Android agent that watches what you do and does it for you. X-OmniClaw scrolls apps, reads prices aloud, and taps through tasks, all on its own. It never leaves your device.
The agent jumps into the shopping app, scrolls, takes screenshots, and reads out prices and sales figures through a vision-language model. A follow-up like "open the second item" works without any extra grounding. In another example, X-OmniClaw acts as a "ScreenAvatar," a "digital surrogate" that solves on-screen tasks on command, like working through a series of practice problems one after another.
A third demo shows the system responding to a request to turn all parrot photos into a highlight album. It gathers matching files, jumps via deeplink into a video editing app's one-click composition tool, and selects the images with multi-tap.
The point isn't the parrot slideshow. It's the method. The agent uses the camera, the screen, your voice.
It needs no special access or APIs, just the same visual and touch inputs a person has. Oppo is giving this away. That means every developer with an idea for an overbearing digital assistant now has a blueprint.
Your phone is watching. Soon it might start doing all the work.
Related Reading
Open Source Trump cracks down on Anthropic after Amazon tip; staff largely foreign
Open Source SDOF Adds Two Defensive Layers via Intent Router and StateAwareDisp
Open Source D&B rebuilds 642 million‑business database after AI agents hit limits
From Archives GraphBit Employs Typed Functions and Rust Engine for Reproducible Orchestration
From Archives Osaurus, an open‑source Mac tool, delivers personal AI with local and cloud models
Common Questions Answered
What capabilities does the Oppo X-OmniClaw agent have on Android devices?
X-OmniClaw is an autonomous agent that can scroll through apps, read prices aloud, and automatically tap through tasks without requiring user intervention. The agent operates entirely on-device, meaning it processes all actions locally on your phone without needing to send data to external servers or cloud services.
How does X-OmniClaw access phone functions without requiring special APIs or permissions?
X-OmniClaw uses the same visual and touch inputs that a person would use, including the camera and screen, to interact with apps and perform tasks. By relying on these standard user-level inputs rather than special system access or APIs, the agent can function across different apps without requiring elevated permissions.
Why is Oppo open-sourcing the X-OmniClaw agent code?
By releasing X-OmniClaw as open-source, Oppo is providing developers with a blueprint and foundation for creating their own autonomous digital assistants. This open approach democratizes access to the technology, enabling any developer to build upon the agent's methods and create customized overbearing digital assistants for various use cases.
What are the privacy implications of having an agent like X-OmniClaw on your phone?
Since X-OmniClaw operates entirely on-device and watches what you do to replicate your actions, it has visibility into your phone activity and the ability to perform tasks autonomously. The agent's camera and screen monitoring capabilities mean your phone is essentially watching and learning your behavior patterns to automate future actions.
Further Reading
- A Unified Mobile Agent for Multimodal Understanding and Interaction — arXiv
- X-OmniClaw Technical Report | Project Page — Project Page
- OPPO-Mente-Lab/X-OmniClaw: an edge-native Multimodal Android Agent that integrates multimodal perception, memory, and action — GitHub
- Repositories - OPPO-Mente-Lab - GitHub — GitHub
- OPPO AI中心大模型算法部 - GitHub — GitHub
