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	14521
title	ADR-19 §2.1: NL/github/gitlab MCP configs participate in the Tier-1 hierarchy (#14499 fix-forward)
state	Closed
labels	bugairegressionarchitecturemodel-experience
assignees	neo-opus-grace
createdAt	Jul 3, 2026, 2:09 AM
updatedAt	Jul 3, 2026, 7:43 AM
githubUrl	https://github.com/neomjs/neo/issues/14521
author	neo-opus-grace
commentsCount	4
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
contentTrust
projected
quarantined	0
signals	[]
blockedBy	[]
blocking	[]
closedAt	Jul 3, 2026, 7:43 AM

gitlab MCP configs participate in the Tier-1 hierarchy (#14499 fix-forward)

Closed v13.1.0/archive-v13-1-0-chunk-8 bugairegressionarchitecturemodel-experience

neo-opus-grace commented on Jul 3, 2026, 2:09 AM

#14499 (Resolves #13432) shipped ADR-19 violations that crashed the live orchestrator and were missed in review by both reviewers (@neo-opus-ada + @neo-opus-grace). Reverting to restore a known-good dev; #13432 will be re-delivered ADR-19-compliant on a clean base.

Damage (V-B-A'd against code, per @neo-opus-ada's diagnosis):

Class 1 — §2.1 hierarchy non-participation (the crash). initServerConfigs.mjs reads aiConfig.auth.mode bare, but NL / non-memory-core server configs do not import the Tier-1 realm root, so their getParent() chain has no root → auth is undefined → TypeError → orchestrator/NL bridge crash. My review RA-1 ("drop the defensive ?. on aiConfig.auth") removed the mask over this non-participation without fixing participation first — sequence-inverted.
Class 2 — §3 B5 pass-along. Daemons thread the SSOT root (assertConfigFresh({aiConfig: AiConfig})) instead of the guard reading the resolved leaf at its own use-site.

Disposition: revert now (stop the live crash) → co-driven ADR-19 audit → compliant #13432 re-do (participation-first, no threading, no B3 ?.) → elevate the AGENTS.md ADR-19 read-gate to CRITICAL. Re-opens #13432.