What is the 'Neural Link'?

The Neural Link is a bi-directional bridge that connects AI agents (like Gemini or Claude) directly to the Neo.mjs runtime. It allows agents to 'see' the application's Scene Graph, inspect component state, verify event listeners, and even mutate the running application in real-time. This turns the application into a 'glass box' for AI, enabling autonomous debugging and feature development.

Why is Neo.mjs called an 'Application Engine' instead of a framework?

Traditional frameworks are general-purpose libraries (like a Toyota) that help you organize code, but they compile away into ephemeral DOM nodes. Neo.mjs is a precision-engineered runtime (like an F1 car), similar to Unreal Engine for games. It maintains a persistent 'Scene Graph' of objects in a separate worker thread. These objects retain their identity, state, and relationships, allowing for advanced capabilities like multi-window orchestration, runtime permutation, and deep AI introspection that are impossible with 'melted plastic' DOM-based frameworks.

What is 'Context Engineering'?

Context Engineering is the practice of curating the environment and information flow for AI agents to maximize their autonomy. In Neo.mjs, this is implemented via three Model Context Protocol (MCP) servers: a Knowledge Base for semantic code understanding, a Memory Core for learning from past sessions, and a GitHub Workflow server for project management. This ecosystem allows agents to work as fully integrated members of the development team.

What is 'Object Permanence' in the context of Neo.mjs?

In Neo.mjs, UI components are persistent JavaScript objects living in the App Worker, not just transient rendering results. This 'Object Permanence' means a component (like a Dashboard) maintains its state (scroll position, user input, internal logic) even if it is detached from the DOM or moved to a different browser window. This is the 'Lego Technic' approach versus the 'Duplo' approach of traditional frameworks.

What is an 'Agent Operating System'?

Neo.mjs v11 introduced the concept of an Agent OS, where the platform itself provides the tools and interfaces for AI agents to operate. It combines a standalone, type-safe AI SDK for autonomous 'Code Execution' with the Neural Link for runtime control. This enables agents to monitor, debug, and heal the application autonomously, effectively acting as an operating system for synthetic intelligence.

How does Neo.mjs handle multi-window applications?

Neo.mjs uses shared web workers to run a single application instance across multiple browser windows. All windows share the same application state and data in real-time. Components can even move between windows while retaining their JavaScript instances. This enables desktop-class experiences like multi-window IDEs, browser-based email clients, and multi-screen control rooms.

What makes Neo.mjs different from React, Angular, or Vue?

Neo.mjs is fundamentally different in its architecture: (1) It uses True Multithreading via web workers to prevent UI jank, (2) It is an AI-Native Application Engine with built-in bridges for AI collaboration, and (3) It treats components as persistent objects in a Scene Graph, enabling native multi-window support and runtime permutation.

Frontmatter

id	9189
title	Enhance Stream Proxy with DecompressionStream Support
state	Closed
labels	enhancementaiperformance
assignees	tobiu
createdAt	Feb 17, 2026, 1:40 AM
updatedAt	Feb 17, 2026, 1:52 AM
githubUrl	https://github.com/neomjs/neo/issues/9189
author	tobiu
commentsCount	1
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Feb 17, 2026, 1:52 AM

Enhance Stream Proxy with DecompressionStream Support

Name: Neo.mjs Application Engine
Author: Neo.mjs

Closed v12.0.0 enhancementaiperformance

tobiu commented on Feb 17, 2026, 1:40 AM

This task enhances the Neo.data.proxy.Stream to support client-side decompression of data streams using the browser's DecompressionStream API. This is a strategic enabler for scaling datasets (e.g., from 50k to 100k+ records) by significantly reducing network transfer size (estimated 70-80% reduction for JSONL).

Objectives:

Proxy Enhancement (src/data/proxy/Stream.mjs):
- Add a compression config ('gzip', 'deflate', or null).
- If set, pipe the fetch response body through new DecompressionStream(this.compression) before reading.
- Ensure compatibility with existing features (Progressive Chunking, Abort).
Proof of Concept (Experiment):
- Manually compress users.jsonl to users.jsonl.gz.
- Configure DevIndex.store.Contributors to load the compressed file with compression: 'gzip'.
- Verify that the app loads correctly and measure the network transfer savings.
Cleanup:
- Revert the store configuration to use the uncompressed file (to maintain meaningful git diffs for data updates).
- Keep the Proxy capability merged.

Acceptance Criteria:

Stream.read() successfully decompresses a .gz file when configured.
Loading a compressed file triggers the same progressive events as an uncompressed one.
Network transfer size is significantly reduced (Validation).
Fallback/Error handling if DecompressionStream is missing (optional but good).

tobiu added the enhancement label on Feb 17, 2026, 1:40 AM

tobiu added the ai label on Feb 17, 2026, 1:40 AM

tobiu added the performance label on Feb 17, 2026, 1:40 AM

tobiu assigned to @tobiu on Feb 17, 2026, 1:41 AM

tobiu referenced in commit beee3f7 - "feat(proxy): Add DecompressionStream support for client-side gzip/deflate handling (#9189)" on Feb 17, 2026, 1:51 AM

tobiu Feb 17, 2026, 1:52 AM

Input from Gemini 3 Pro:

✦ I have implemented DecompressionStream support in Neo.data.proxy.Stream.

Features:

Config: compression: 'gzip' | 'deflate' | null.

Implementation: Transparently pipes the fetch stream through the browser's native DecompressionStream.

Impact: Enables serving .jsonl.gz files directly.

Proof of Concept: Validated a ~60% size reduction (14MB -> 5.8MB) for the DevIndex dataset with full functionality.

This enhancement, combined with Progressive Chunk Sizing (#9187) and Abort support (#9188), positions the data layer to handle 100k+ record datasets efficiently.

tobiu closed this issue on Feb 17, 2026, 1:52 AM