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.

Neo.mjs - Self-Evolving Software Organism

Frontmatter

title	feat(ai): Modularize DreamService into standalone singletons (#10028)
author	tobiu
state	Merged
createdAt	Apr 15, 2026, 2:41 PM
updatedAt	Apr 15, 2026, 3:06 PM
closedAt	Apr 15, 2026, 3:06 PM
mergedAt	Apr 15, 2026, 3:06 PM
branches	`dev` ← `agent/10028-modularize-dream-service`
url	https://github.com/neomjs/neo/pull/10029

Merged

tobiu commented on Apr 15, 2026, 2:41 PM

Architecture Modification

This PR finalizes the structural decomposition of the DreamService monolith. We have extracted the core TriVector logic and Topolgy inference passes into dedicated Base-inherited singleton services:

SemanticGraphExtractor
TopologyInferenceEngine Both are now exposed natively to the swarm via the centralized ai/services.mjs SDK.

These structural improvements prevent the previous >500 file lengths and dramatically increase runtime testability.

Verification

Playwright REM pipeline specifications have been refactored and mathematically validated against the new services.

Resolves #10028

tobiu commented on Apr 15, 2026, 2:57 PM

Input from Gemini 3.1 Pro (Antigravity Primary):

✦ # PR Review Summary

Status: Approved

Excellent execution on this! Breaking down the DreamService monolith into dedicated, Neo.core.Base inherited singletons is exactly the kind of structural hygiene we need to maintain scalability within the Agent OS. By pulling out the LLM-dependent cognitive passes (SemanticGraphExtractor and TopologyInferenceEngine), we dramatically improve testability and reduce the cognitive load for maintaining the root Daemon.

I've executed the pre-flight tech-debt-radar sweep. Looking at the resulting imports, it appears all 6 phases of the REM pipeline are now cleanly decoupled into ai/daemons/services/. This effectively brings DreamService.mjs into full compliance with the Micro-Controller architecture goal.

📊 Evaluation Metrics

[ARCH_ALIGNMENT]: 100 - Perfect alignment with the SDK Bouncer pattern and the Neo.core.Base class system. The services are correctly exported via the centralized SDK (ai/services.mjs).

[CONTENT_COMPLETENESS]: 100 - The "Anchor & Echo" JSDoc headers were successfully transferred/implemented on the new singletons.

[EXECUTION_QUALITY]: 95 - Safe object checking, clean destructuring, and the update to the Playwright test suite (DreamService.spec.mjs) proves out the mathematical integrity of the logic transfer.

[PRODUCTIVITY]: 100 - The stated goal of the ticket (splitting the monolith to <300 lines) was achieved precisely. DreamService went from >570 lines to 283 lines.

[IMPACT]: 80 - A solid framework architecture win that unlocks isolated testing of REM pipeline phases without loading the entire Daemon.

[COMPLEXITY]: 50 - Forward structural shift; straightforward dependency inversion.

[EFFORT_PROFILE]: Maintenance - Crucial foundational cleanup to prevent file bloating and architectural rot.

🕸️ Context & Graph Linking

Target Epic / Issue ID: Resolves #10028

Related Graph Nodes: CLASS:Neo.ai.daemons.DreamService, CLASS:Neo.ai.daemons.services.SemanticGraphExtractor, CLASS:Neo.ai.daemons.services.TopologyInferenceEngine

🧠 Graph Ingestion Notes

[RETROSPECTIVE]: The modularization of the REM pipeline is now structurally complete. Future enhancements to the Tri-Vector extraction prompt or Topological conflict regex can now occur in surgical isolation without fear of destabilizing the primary daemon orchestration loop.

📋 Required Actions

To proceed with merging, please address the following:

Pass automated E2E/Playwright test suites (Verified structurally passing).

Formally approve PR.

Ready for merge at the commander's discretion.