What is the Neural Link?

The Neural Link is a bi-directional bridge that connects AI agents directly to the Neo.mjs runtime. It lets agents inspect the Scene Graph, component state, event listeners, computed styles, and DOM rectangles, and mutate the running application in real time.

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

Neo.mjs maintains persistent application objects in a worker-backed Scene Graph instead of compiling application state away into ephemeral DOM nodes. That architecture enables multi-window orchestration, runtime permutation, and deep AI introspection.

What is Context Engineering?

Context Engineering shapes the information and tool environment around AI agents. Neo.mjs implements it through Knowledge Base, Memory Core, GitHub Workflow, and Neural Link MCP servers for frontier harnesses, plus a File System MCP server for internal Neo.ai.Agent local loops.

What is the Neo.mjs Agent OS?

The Neo.mjs Agent OS is the repository Brain: source code and services for Memory Core, Knowledge Base, Active Hybrid GraphRAG, DreamService, Golden Path synthesis, A2A coordination, and Neural Link tooling.

Frontmatter

id	10014
title	[Sub-Epic] Macro Knowledge Base & Semantic Gap Inference
state	Closed
labels	epicaiarchitecture
assignees	[]
createdAt	Apr 14, 2026, 6:43 PM
updatedAt	Apr 16, 2026, 8:44 PM
githubUrl	https://github.com/neomjs/neo/issues/10014
author	tobiu
commentsCount	1
parentIssue	null
subIssues	10003 Standardize Global Vector Embedding Configuration 10002 Universal Macro Overlay DB Topology 10004 DreamService: Omnidirectional Gap Inference Engine 10006 DreamService: Semantic Edge Injection into Native Graph 10012 DreamService: Federated Execution Batching & Throttling
subIssuesCompleted	5
subIssuesTotal	5
contentTrust
projected
quarantined	0
signals	[]
blockedBy	[x] 10028 Finalize DreamService Decomposition: Tri-Vector & Topology Extractors, [x] 10013 [Sub-Epic] DreamService Decomposition
blocking	[]
closedAt	Apr 16, 2026, 8:44 PM

[Sub-Epic] Macro Knowledge Base & Semantic Gap Inference

Closed v13.0.0/archive-v13-0-0-chunk-4 epicaiarchitecture

tobiu commented on Apr 14, 2026, 6:43 PM

Objective

Build an unchunked "Macro" embedding collection alongside the existing chunked Knowledge Base, enabling mathematically meaningful full-file semantic distance comparisons between source files, guides, tickets, and tests.

Rationale

The current executeCapabilityGapInference() uses RegExp token matching against file paths. This fails for conceptual guides (e.g., Reactivity.md never token-matches Neo.button.Base). Full-file vector embeddings solve this by enabling direct semantic distance measurement between any two artifacts in the project.

Scope

This sub-epic groups the following existing tickets:

#10003 — Standardize Global Vector Embedding Configuration
#10002 — Universal Macro Overlay DB Topology
#10004 — DreamService: Omnidirectional Gap Inference Engine
#10006 — DreamService: Semantic Edge Injection into Native Graph
#10012 — DreamService: Federated Execution Batching & Throttling

Key Constraint

The embedding model used for the Macro DB MUST be identical to the one used by the Memory Core. This is enforced by NEO_GLOBAL_EMBEDDING (#10003). Model changes require a full re-embed of all collections.

Parent Epic: #9999 Origin Session ID: 69b726ca-97c8-406c-9106-c63b6642c4c0

tobiu cross-referenced by #10030 on Apr 15, 2026, 11:55 PM

tobiu Apr 16, 2026, 8:44 PM

Superseded by Epic #10030. The vector-based Macro Overlay topology has been abandoned in favor of the lightweight JSONL Concept Graph approach to solve semantic dilution.

tobiu cross-referenced by #9999 on Apr 16, 2026, 8:44 PM

tobiu cross-referenced by PR #10047 on Apr 17, 2026, 9:38 AM

tobiu cross-referenced by PR #10066 on Apr 18, 2026, 8:15 PM