AirJelly launches a desktop proactive AI assistant, using the Enter key as an anchor to capture user intent

robot
Abstract generation in progress
ME News Report, April 22 (UTC+8), according to Beating Monitoring, the persistent low-entropy (Low Entropy AI) product founded by former ByteDance product manager Bert has launched a desktop AI assistant AirJelly and has secured funding from Source Capital. Bert previously led the context engineering product MineContext at ByteDance, then left with his original team to start a business, consisting of 11 members, all working offline in Beijing.
AirJelly does not perform full-screen recording but uses the Enter key as an anchor: each time the user presses Enter, it captures the screen, and through Accessibility permissions, it obtains the current application, input box type, and context, modeling this information as an Event, which is then summarized into a Task by AI.
Compared to the previous MineContext's full-screen approach with about 1,500 screenshots daily, the Enter mechanism reduces the screenshot volume to around 300, cutting costs to one-fifth and also decreasing false positives caused by irrelevant screenshots.
The memory system has two layers: static information is modeled as Entities (people, projects, etc.), dynamic information as Tasks, each containing a title, summary, progress, next steps, and related Events. During recall, it combines vector retrieval, keyword search, and time decay weighting, with all data stored locally.
Based on this, AirJelly determines whether the Task update reaches the push threshold and proactively recommends the next action to the user, which the team calls Proactive Trigger.
The underlying agent execution capability is integrated with OpenClaw’s Pi framework, combined with MineContext’s screen understanding ability.
The product currently supports macOS, with Windows and Linux versions in development, available for free download at airjelly.ai.
The team’s next plan is to launch the "Next Enter Prediction" feature, which predicts the content of the next Enter input based on historical behavior trajectories, and to support a team version for collaborative work.
(Source: BlockBeats)
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned