TECNO has announced the next stage of its EllaClaw mobile AI agent, expanding it from basic system tasks into cross-app automation and deeper device management. The update gives the agent the ability to work across third-party apps including shopping, food delivery, and ride-hailing platforms, while also handling phone-level tasks like clearing RAM, monitoring battery drain, and managing mobile data usage.
EllaClaw is still in closed beta and aimed squarely at users in emerging markets, where TECNO has the bulk of its audience. The company says the agent runs in the cloud and operates in the background, with a confirmation-first policy that requires user approval before making significant changes to the device. That kind of transparency matters in a space where many AI features feel opaque or hard to trust.
The timing fits a broader shift happening across the mobile industry. Smartphone makers and software developers are moving away from chatbot-style AI, which simply answers questions, toward agentic AI that actually does things on your behalf. Apple, Google, and Samsung have all been pushing in this direction, and TECNO’s move with EllaClaw shows that the trend is spreading well beyond the premium segment.
On the device management side, EllaClaw draws on more than 40 built-in skills. The most relevant ones for everyday users include:
- Smart CleanUp Boost, which frees up RAM and CPU when the phone starts to slow down
- Smart Power Drain Check, which identifies apps consuming excessive battery
- Instant Cool-down Relief, which reduces background activity during demanding tasks
- Smart Data Guardian, which tracks mobile data use against personal habits to help avoid overage
The cross-app features are where things get more ambitious. With user permission, EllaClaw can interact with third-party apps by reading their interfaces visually, the same way a person would, rather than tapping into hidden APIs. TECNO calls this GUI comprehension. It means users can watch the agent navigate an app step by step, rather than having to trust something happening invisibly in the background. Supported categories currently include shopping platforms like Lazada, transportation apps, food delivery, and smart home controls.
The agent also has a memory layer that builds up over time. It learns recurring habits and preferences, which lets it do things like prepare a morning briefing that pulls together calendar events, weather, and news, or act as a trip assistant that books a ride, sets a departure alarm, and organizes itinerary details. TECNO also mentions reminders to contact family members based on schedules and location data, which is a small but telling detail about who the target user is.
Jack Guo, General Manager of TECNO, said the company’s aim is to create AI that reduces friction and stays transparent. “Our goal is to create AI that helps simplify everyday tasks, reduce friction and make advanced experiences more accessible, while ensuring transparency and user control remain central to the experience,” he said.
EllaClaw is not a finished product. TECNO describes it as an exploratory concept in internal testing and closed beta, and has not announced a public release date or specific market rollout. That caution is reasonable. Agentic AI that can control apps and modify system settings carries real risk if it gets things wrong, and the confirmation-first approach suggests TECNO is aware of that. How well the guardrails hold up in real-world use will be the real test when it eventually ships more broadly.
For now, EllaClaw is an interesting signal about where the mid-range and emerging-market smartphone segment is heading. If the agent works as described, it could bring capabilities to users who have largely been excluded from the AI feature race, not because the technology was unavailable, but because no one had adapted it for their context.
