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	7949
title	Refactor Shared Tool Validation Service for MCP Client and Server
state	Closed
labels	enhancementairefactoring
assignees	tobiu
createdAt	Nov 30, 2025, 2:50 PM
updatedAt	Nov 30, 2025, 3:29 PM
githubUrl	https://github.com/neomjs/neo/issues/7949
author	tobiu
commentsCount	1
parentIssue	7931
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Nov 30, 2025, 3:29 PM

Refactor Shared Tool Validation Service for MCP Client and Server

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

Closed v11.14.0 enhancementairefactoring

tobiu commented on Nov 30, 2025, 2:50 PM

Context

We implemented basic client-side validation in #7946. However, the validation logic is currently fragmented. We want to unify it by moving the tool service to a shared location.

Goal

Refactor ai/mcp/server/toolService.mjs to be a shared resource for both Clients and Servers.

Requirements

Relocate File: Move ai/mcp/server/toolService.mjs one folder level up (to ai/mcp/toolService.mjs) to allow shared access.
Add validateToolInput(): Implement a shared method validateToolInput(toolName, args, schema) within the shared service.
Refactor Client & Server: Update both Neo.ai.mcp.client.Client and the MCP Servers to import and use this shared service for validation.
Server Type Strategy:
- Implement a check to determine if the connected server is one of our internal servers (GitHub, Knowledge Base, Memory Core).
- Ensure validation logic is applied appropriately for internal servers (using our shared logic). For external servers, we need to decide if we validate against the provided JSON schema or skip strict validation.

Deliverables

Shared ai/mcp/toolService.mjs.
Updated Client.mjs and Server implementations.

tobiu added the enhancement label on Nov 30, 2025, 2:50 PM

tobiu added the ai label on Nov 30, 2025, 2:50 PM

tobiu added the refactoring label on Nov 30, 2025, 2:50 PM

tobiu assigned to @tobiu on Nov 30, 2025, 2:51 PM

tobiu added parent issue #7931 on Nov 30, 2025, 2:51 PM

tobiu referenced in commit 4ea7395 - "Refactor Shared Tool Validation Service for MCP Client and Server #7949" on Nov 30, 2025, 3:28 PM

tobiu Nov 30, 2025, 3:29 PM

Input from Gemini 2.5 pro:

✦ I have successfully completed the refactoring for issue #7949.

Achievements

Shared Service: Relocated toolService.mjs to ai/mcp/toolService.mjs, making it accessible to both Server and Client components.

Hybrid Validation: Implemented validateToolInput(toolName, args, schema) which supports:

Internal Zod Validation: Uses robust Zod schemas derived from openapi.yaml (when initialized, e.g., in Servers).

External JSON Schema Validation: Falls back to a lightweight JSON Schema validator (when Zod schemas are unavailable, e.g., in Clients connecting to external servers).

Integration:

Updated Neo.ai.mcp.client.Client to use the shared validateToolInput.

Updated github-workflow, knowledge-base, and memory-core servers to use the shared service.

Future Work (Singleton Refactoring)

We identified that toolService.mjs is currently a singleton module. This prevents a single process (like an Agent) from initializing multiple internal server contexts (Zod schemas) simultaneously.

Current Impact: Client.mjs uses the JSON Schema fallback for validation, which is functional and safe.

Next Step: A follow-up ticket will be created to refactor toolService.mjs into a Class-based service to support multiple instances per process, enabling Zod validation on the Client side for multiple internal servers.

I am closing this ticket as the primary goal of sharing the validation logic and structure is complete.

tobiu closed this issue on Nov 30, 2025, 3:29 PM

tobiu cross-referenced by #7950 on Nov 30, 2025, 3:38 PM