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	8169
title	Neural Link Core Capabilities
state	Closed
labels	epicaiarchitecture
assignees	tobiu
createdAt	Dec 28, 2025, 2:36 PM
updatedAt	Jan 5, 2026, 5:05 PM
githubUrl	https://github.com/neomjs/neo/issues/8169
author	tobiu
commentsCount	0
parentIssue	null
subIssues	8170 Implement Neural Link Client Core (Tree & Props) 8171 Implement Neural Link Window Topology Discovery 8172 Implement Neural Link Drag & Drop Inspection 8016 Harden Neural Link MCP Server 8173 Implement Unique App Worker Identification & Worker Topology 8174 Refactor Neural Link Routing: Standardize Session vs Window ID 8175 Implement Window Connect/Disconnect Notifications for Neural Link 8176 Implement Neural Link State Rehydration on Reconnect 8177 Harden WebSocket Connection: Backoff & Auto-Reconnect 8178 Refactor: Extract ComponentService from ConnectionService 8179 Refactor: Extract RuntimeService from ConnectionService 8180 Refactor: Extract InteractionService from ConnectionService 8181 Feat: ConnectionService Lifecycle Tools 8183 Feat: Neural Link - Data Store Inspection & Manager 8184 Feat: Neural Link - Navigation Control & History 8185 Feat: Neural Link - Visual Inspection (DomRect) 8186 Feat: Neural Link - Semantic Component Query 8187 Feat: Neural Link - State Provider Inspection & Modification 8188 Feat: Neural Link - Visual Highlighting 8189 Feat: Neural Link - Advanced Event Simulation 8190 Feat: Neural Link - Runtime Code Hot-Patching 8191 Feat: Neural Link - Global Config Management 8192 Feat: Neural Link - Log & Error Streaming 8193 Feat: Neural Link - Namespace Discovery 8182 Feat: Neural Link Identity and Recovery 8194 Feat: Neural Link - Propagate Runtime Errors to Agent 8195 Feat: Neural Link - Get Record Tool 8196 Refactor: Neural Link - Client-Side Service Architecture 8197 Feat: Neural Link - Export Domain Services in AI SDK 8206 Refactor Client Services to use Parameter Destructuring 8207 ai.Client: Enhance handleRequest to support sync and async execution 8208 ai.client.ComponentService: Make highlightComponent non-blocking 8210 Create standalone Neural Link Bridge process 8211 Update ConnectionService to use Neural Link Bridge 8209 Fix silent failure in Neural Link ConnectionService startup 8212 Enhance Neural Link Health Check with detailed diagnostics 8213 Log App Name in Neural Link Bridge connections 8214 Fix Neural Link Server startup crash due to health check schema mismatch 8215 Expose connected agents in Neural Link health check 8216 Fix Neural Link server startup crash due to missing startServer/stopServer methods 8222 Fix Neural Link Recovery: Exempt server control tools from health check gate 8223 Investigate and Fix get_component_tree Timeout 8271 [Neural Link] Refactor ComponentService to use toJSON protocol 8272 [Neural Link] Refactor DataService to use toJSON protocol 8273 [Neural Link] Refactor RuntimeService to use toJSON protocol 8274 [Neural Link] Parameterize inspectStore limit and offset 8278 [Neural Link] Feature: Tool get_dom_event_listeners 8279 [Neural Link] Feature: Tool get_dom_event_summary 8280 [Neural Link] Feature: Tool inspect_class 8281 [Neural Link] Feature: Tool get_computed_styles 8282 [Neural Link] Enhance inspect_class tool description 8283 [Neural Link] Feature: inspect_class tiered detail (compact mode) 8286 Enhance DomAccess.getElement to support window and document targets 8305 Feat: Neural Link - Buffer Early Logs 8307 Neural Link: Support depth in getVdomTree and getVnodeTree 8308 Neural Link: Implement atomic get_vdom_and_vnode tool 8309 Feat: Neural Link - Get Method Source 8310 Feat: Environment-Aware Neural Link Config 8311 Feat: Neural Link - Configurable URL & Graceful Connection 8312 Fix Neural Link Window Topology Race Condition 8313 Fix Neural Link Server startup handshake deadlock 8322 Feat: Make Neural Link RPC timeout configurable 8323 Fix Neural Link Tool Mappings 8324 [Neural Link] Generalize Property Access to InstanceService 8325 [Neural Link] Enhance Instance Tools with Batch Operations 8326 [Neural Link] Enhance query_component with Property Return 8327 [Neural Link] Feature: Tool find_instances 8328 [Neural Link] Feature: Tool query_vdom 8329 [Neural Link] Refactor: ComponentService.queryComponent logic 8330 [Neural Link] Fix find_instances regression (Client mapping & OpenAPI) 8331 Enhance Neural Link Health Check Response 8332 Fix manage_connection tool argument passing 8333 Fix Neural Link Server startup crash due to Health Check response structure change 8334 Fix query_component missing returnProperties argument
subIssuesCompleted	74
subIssuesTotal	74
blockedBy	[]
blocking	[]
closedAt	Jan 5, 2026, 5:05 PM

Neural Link Core Capabilities

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

Closed v11.18.0 epicaiarchitecture

tobiu commented on Dec 28, 2025, 2:36 PM

This epic tracks the implementation of core capabilities for the Neural Link, enabling AI agents to inspect and modify the running Neo.mjs application state via the neo-neural-link MCP server.

Goal: Empower AI agents to "see" and "touch" the runtime application, moving beyond static code analysis. This is a prerequisite for advanced features like "Infinite Canvas" debugging, "Self-Healing Apps," and "Runtime Orchestration."

Scope:

Read/Write Primitives: Implement the ability to retrieve component/vdom/vnode trees and get/set properties via Neo.ai.Client.
Runtime Inspection Tools: Create specialized MCP tools to inspect complex internal subsystems (Drag & Drop, Focus, etc.).
Topology Discovery: Enable agents to map windowIds to logical application parts (Main Window, Popups) with rich metadata.

Success Criteria:

An AI agent can query the current component tree (with configurable depth and serialization safety).
An AI agent can read and write properties/configs of a live component.
An AI agent can inspect the state of a drag operation in real-time.

tobiu added the epic label on Dec 28, 2025, 2:36 PM

tobiu added the ai label on Dec 28, 2025, 2:36 PM

tobiu added the architecture label on Dec 28, 2025, 2:36 PM

tobiu assigned to @tobiu on Dec 28, 2025, 2:36 PM

tobiu added sub-issue #8170 on Dec 28, 2025, 2:37 PM

tobiu added sub-issue #8171 on Dec 28, 2025, 2:37 PM

tobiu added sub-issue #8172 on Dec 28, 2025, 2:38 PM

tobiu added sub-issue #8016 on Dec 28, 2025, 2:39 PM

tobiu added sub-issue #8173 on Dec 28, 2025, 3:25 PM

tobiu added sub-issue #8174 on Dec 28, 2025, 4:40 PM

tobiu added sub-issue #8175 on Dec 28, 2025, 6:39 PM

tobiu added sub-issue #8176 on Dec 28, 2025, 6:39 PM

tobiu added sub-issue #8177 on Dec 28, 2025, 6:58 PM

tobiu referenced in commit 21b8247 - "feat(ai): Implement Neural Link healing and standardize routing (#8169) on Dec 28, 2025, 7:15 PM

tobiu added sub-issue #8178 on Dec 28, 2025, 7:42 PM

tobiu added sub-issue #8179 on Dec 28, 2025, 7:42 PM

tobiu added sub-issue #8180 on Dec 28, 2025, 7:42 PM

tobiu added sub-issue #8181 on Dec 28, 2025, 7:42 PM

tobiu added sub-issue #8183 on Dec 28, 2025, 9:55 PM

tobiu added sub-issue #8184 on Dec 28, 2025, 9:59 PM

tobiu added sub-issue #8185 on Dec 28, 2025, 10:03 PM

tobiu added sub-issue #8186 on Dec 28, 2025, 10:06 PM

tobiu added sub-issue #8187 on Dec 28, 2025, 10:12 PM

tobiu added sub-issue #8188 on Dec 28, 2025, 10:16 PM

tobiu added sub-issue #8189 on Dec 28, 2025, 10:19 PM

tobiu added sub-issue #8190 on Dec 28, 2025, 10:26 PM

tobiu added sub-issue #8191 on Dec 28, 2025, 10:30 PM

tobiu added sub-issue #8192 on Dec 28, 2025, 10:32 PM

tobiu added sub-issue #8193 on Dec 28, 2025, 10:37 PM

tobiu added sub-issue #8182 on Dec 28, 2025, 10:40 PM

tobiu added sub-issue #8194 on Dec 28, 2025, 11:18 PM

tobiu added sub-issue #8195 on Dec 28, 2025, 11:43 PM

tobiu added sub-issue #8196 on Dec 29, 2025, 12:40 AM

tobiu added sub-issue #8197 on Dec 29, 2025, 1:03 AM

tobiu added sub-issue #8206 on Dec 30, 2025, 12:27 AM

tobiu added sub-issue #8207 on Dec 30, 2025, 12:59 AM

tobiu added sub-issue #8208 on Dec 30, 2025, 1:16 AM

tobiu added sub-issue #8210 on Dec 30, 2025, 10:18 AM

tobiu added sub-issue #8211 on Dec 30, 2025, 10:18 AM

tobiu added sub-issue #8209 on Dec 30, 2025, 10:18 AM

tobiu added sub-issue #8212 on Dec 30, 2025, 10:46 AM

tobiu added sub-issue #8213 on Dec 30, 2025, 11:00 AM

tobiu added sub-issue #8214 on Dec 30, 2025, 11:19 AM

tobiu added sub-issue #8215 on Dec 30, 2025, 11:24 AM

tobiu added sub-issue #8216 on Dec 30, 2025, 12:14 PM

tobiu added sub-issue #8222 on Dec 30, 2025, 7:46 PM

tobiu added sub-issue #8223 on Dec 30, 2025, 7:49 PM

tobiu added sub-issue #8271 on Jan 1, 2026, 6:27 PM

tobiu added sub-issue #8272 on Jan 1, 2026, 6:27 PM

tobiu added sub-issue #8273 on Jan 1, 2026, 6:27 PM