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	13202
title	FleetManager seams: plain fields matching the FleetLifecycleService precedent (#13194 review follow-up)
state	Closed
labels	airefactoringarchitecture
assignees	neo-opus-ada
createdAt	Jun 14, 2026, 10:18 AM
updatedAt	Jun 21, 2026, 2:17 AM
githubUrl	https://github.com/neomjs/neo/issues/13202
author	neo-opus-ada
commentsCount	0
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Jun 14, 2026, 1:59 PM

FleetManager seams: plain fields matching the FleetLifecycleService precedent (#13194 review follow-up)

Closed v13.1.0/archive-v13-1-0-chunk-2 airefactoringarchitecture

neo-opus-ada commented on Jun 14, 2026, 10:18 AM

Context

#13194 (FleetManager facade) merged with its injectable test seams (lifecycleService, provisionAndStartFn, repoStatusFn) + managedRoot in static config (reactive). @neo-opus-grace's review flagged the divergence (non-blocking comment, PRR_kwDODSospM8AAAABC8eYsw): the sibling FleetLifecycleService (the established precedent) keeps its tunables (credentialEnvVar / bridgeTokenEnvVar) and seams (registry / spawnFn) as plain class fields, with only className / singleton in static config. The PR merged before the fix landed, so this is the follow-up.

The Problem

The seams are test doubles — nothing observes or binds them — so reactive config adds no value, diverges from the sibling shape, and risks the AiConfig-singleton test-mutation bleed class (the §aiconfig_ssot / ADR 0019 B4 concern). V-B-A confirmed the precedent: FleetLifecycleService :76/:83/:103/:110 are all plain fields.

The Fix

ai/services/fleet/FleetManager.mjs: move managedRoot + the three seams from static config to plain class fields (= null), leaving only className / singleton in config — matching FleetLifecycleService. No behavior/API change: getManagedRoot / getLifecycleService / etc. read the same this.<field>; the spec mutates the fields identically (already verified 6/6 green on the change).

Acceptance Criteria

FleetManager static config holds only className + singleton; managedRoot + the 3 seams are plain fields.
FleetManager.spec stays green (6/6) with no test change needed.
node --check OK.

Out of Scope

Any behavior/API change (pure shape refactor).

Follow-up to #13192 / the merged #13194; the FleetLifecycleService plain-field precedent.

Live latest-open sweep: checked immediately before filing; no equivalent.

Authored by Claude Opus 4.8 (Claude Code), @neo-opus-ada. Surfaced by @neo-opus-grace's #13194 review. Origin Session ID: 4c598c8f-d8a7-4288-9420-e825a45d310e