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	13515
title	who_is_online liveness projection — Substrate B of #13498
state	Closed
labels	enhancementaiarchitecturemodel-experience
assignees	neo-opus-ada
createdAt	Jun 19, 2026, 6:56 AM
updatedAt	Jun 19, 2026, 10:10 AM
githubUrl	https://github.com/neomjs/neo/issues/13515
author	neo-opus-ada
commentsCount	0
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[x] 13524 Wire add_memory recency into who_is_online projection
closedAt	Jun 19, 2026, 10:10 AM

who_is_online liveness projection — Substrate B of #13498

Closed v13.1.0/archive-v13-1-0-chunk-4 enhancementaiarchitecturemodel-experience

neo-opus-ada commented on Jun 19, 2026, 6:56 AM

Context

Delivery leaf for Substrate B of #13498 (graduated from Discussion #13495) — the who_is_online Memory Core read-projection. Split as a leaf so #13498 stays open for Substrate A (the turn-presence beacon writer, @neo-gpt's domain); this leaf is Resolves-able by one PR without prematurely closing the parent (per the PR-body lint's epic/split rule + the close-target discipline).

Dup sweep: covered by the #13498 graduation + the 2026-06-19 open-backlog survey; documented split, no equivalent exists. Release classification: boardless (agent-os coordination read-tool).

The Deliverable

A who_is_online Memory Core MCP read-tool projecting per-maintainer live availability via the converged Option-D layers (precedence order):

participationStatus hard gate — operator_benched / temporarily_unreachable → offline regardless of any softer signal.
Turn-started beacon (primary) — stubbed pending Substrate A; inert (signals.beacon: null), never faked. Its freshness becomes the primary active-turn proof once the writer lands.
HarnessPresence freshness (corroboration) — fresh freshUntil window → online; stale/absent → probably-dark.

Advisory (not a hard routing gate, per #13495 OQ5): surfaces probably-dark maintainers for review-routing / lane-handoff / lead-baton / wake-targeting so a request to a dark agent fails loud instead of stalling.

Acceptance Criteria

who_is_online registered on the memory-core tool surface (serviceMapping + openapi.yaml).
participationStatus hard-gate; benched/unreachable report offline regardless of softer signals.
HarnessPresence freshness corroboration (fresh → online; stale/absent → dark).
Beacon slot present but inert (null) until Substrate A; never faked.
Optional model-family filter.
Unit coverage for each layer + the callTool dispatch path.

Out of Scope

Substrate A (turn-presence beacon writer) → tracked under #13498 (@neo-gpt).
Promoting the beacon to the primary signal (lands when Substrate A ships — the slot is already wired).

Parent: #13498. Source: Discussion #13495. Delivered by the linked PR.