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	8066
title	Standardize Renderer Component Lifecycle Management
state	Closed
labels	airefactoringarchitecture
assignees	tobiu
createdAt	Dec 9, 2025, 2:48 AM
updatedAt	Dec 9, 2025, 3:54 AM
githubUrl	https://github.com/neomjs/neo/issues/8066
author	tobiu
commentsCount	1
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Dec 9, 2025, 3:54 AM

Standardize Renderer Component Lifecycle Management

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

Closed v11.17.0 airefactoringarchitecture

tobiu commented on Dec 9, 2025, 2:48 AM

Refactor the management of child components generated by renderers to ensure consistent lifecycle handling (mounting/destroying) across different contexts (LivePreview vs ContentComponent).

Problem: ContentComponent manually manages the mounted state of Markdown-embedded components (livePreviews, customComponents) to support re-mounting (e.g., navigating between portal tabs). This logic is tightly coupled to the container and not reusable by LivePreview when running in markdown mode.

Solution: Shift the responsibility of managing these "floating" components to the renderer that creates them.

Changes:

src/code/renderer/Markdown.mjs:
- Add activeComponents array to track created instances.
- Implement updateComponentState(mounted): Iterates activeComponents and sets their mounted config.
- Implement destroyComponents(): Destroys all activeComponents and clears the array.
- Update render() to populate activeComponents instead of returning them.
src/code/renderer/Base.mjs:
- Define abstract updateComponentState(mounted) (no-op by default).
- Define abstract destroyComponents() (no-op by default).
src/code/LivePreview.mjs:
- Remove customComponents and livePreviews.
- Implement afterSetMounted(value): Call this.renderer?.updateComponentState(value).
- Update destroyChildInstances: Call this.renderer?.destroyComponents().
apps/portal/view/learn/ContentComponent.mjs:
- Remove customComponents and livePreviews.
- Update afterSetMounted: Delegate to this.renderer?.updateComponentState(value).
- Update destroyChildInstances: Delegate to this.renderer?.destroyComponents().

Goal: Decouple component lifecycle management from the container view, ensuring LivePreview and ContentComponent can both correctly handle re-mounting scenarios for Markdown content.

tobiu added the ai label on Dec 9, 2025, 2:48 AM

tobiu added the refactoring label on Dec 9, 2025, 2:48 AM

tobiu added the architecture label on Dec 9, 2025, 2:48 AM

tobiu assigned to @tobiu on Dec 9, 2025, 3:41 AM

tobiu referenced in commit 8623d74 - "Standardize Renderer Component Lifecycle Management #8066" on Dec 9, 2025, 3:43 AM

tobiu Dec 9, 2025, 3:44 AM

Input from Gemini:

✦ I have verified the rationale and implementation of the mounted state handling in ContentComponent and MarkdownRenderer.

Analysis: The regression was caused by the initial mounted setter call during construction (where value is false and oldValue is undefined), which prematurely triggered updateComponentState. Additionally, simply setting component.mounted = true was insufficient for re-mounting components that were removed from the DOM but kept in memory.

Fixes Implemented:

ContentComponent.afterSetMounted: Added if (me.isConstructed) check. This prevents the renderer update from running during the initial config application, ensuring it only responds to actual lifecycle changes (like tab activation/deactivation).

MarkdownRenderer.updateComponentState:

For mounted === true: Changed component.mounted = true to component.initVnode(true). This forces a re-initialization of the VNode tree and a fresh mount, ensuring the component correctly re-attaches to the DOM with its current state.

For mounted === false: Kept component.mounted = false. Crucially, this does NOT trigger a DOM unmount itself (the parent container handles the DOM removal). This step simply updates the internal state flag of the child component to reflect that it is no longer in the document, ensuring its state is consistent for the next mount cycle.

Conclusion: The changes correctly address the lifecycle synchronization issue. The use of initVnode(true) is the correct approach for re-mounting components that have been detached but preserved in memory, especially in the context of the Portal app's card layout where DOM nodes are removed but JS instances persist.

tobiu closed this issue on Dec 9, 2025, 3:54 AM