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	14328
title	docs(agentos): author the A2A messaging guide — the swarm''s nervous system (local wakes vs cloud turn-start)
state	Closed
labels	documentationaiarchitecture
assignees	neo-gpt
createdAt	Jun 29, 2026, 11:46 AM
updatedAt	Jun 30, 2026, 3:53 AM
githubUrl	https://github.com/neomjs/neo/issues/14328
author	neo-opus-grace
commentsCount	0
parentIssue	14310
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
contentTrust
projected
quarantined	0
signals	[]
blockedBy	[]
blocking	[]
closedAt	Jun 30, 2026, 3:53 AM
milestone	v13.1

docs(agentos): author the A2A messaging guide — the swarm's nervous system (local wakes vs cloud turn-start)

Closed v13.1.0/archive-v13-1-0-chunk-8 documentationaiarchitecture

neo-opus-grace commented on Jun 29, 2026, 11:46 AM

Part of #14310 · content rubric + §Mandatory V-B-A grounding in #14310. Refs #14310.

Problem: A2A (agent-to-agent messaging) is a hot topic most AI tech lacks — and Neo has no guide for it. It's a core differentiator: durable messages + wake events + peer-readable reasoning ("telepathy") across rival model families.

Acceptance Criteria

New guide on A2A: durable messaging, the mailbox, peer-readable thought trails (telepathy), cross-family reasoning reads.
The local-vs-cloud delivery model, crisp: LOCAL = wake-up prompts + heartbeats + the wake daemon; CLOUD = agents read messages at the start of every turn (works WITHOUT wakes).
problem → resolve → why-care → what's-in-it; lead with the differentiator (most AI has no A2A); honest.
Registered in tree.json (Brain / coordination area).

⚠️ V-B-A grounding (mandatory before writing): ai/services/memory-core (MailboxService + the A2A primitives: add_message/list_messages/wake), ai/daemons/wake (the wake daemon), the turn-start message-read path (how cloud agents read at turn start), ADR-0002 (wake substrate). Write from VERIFIED reality; capture grounding evidence in the PR.

Files: new learn/agentos/A2A.md (name TBD) + learn/tree.json.

tobiu closed this issue on Jun 30, 2026, 3:53 AM

tobiu referenced in commit 26a0fae - "docs(agentos): add A2A messaging guide (#14328) (#14374)" on Jun 30, 2026, 3:53 AM