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	10318
title	chore(ai): empirical token-economy anchor for swarm heartbeat
state	Closed
labels	documentationenhancementaiarchitectureperformance
assignees	neo-gpt
createdAt	Apr 25, 2026, 4:53 AM
updatedAt	May 1, 2026, 8:29 PM
githubUrl	https://github.com/neomjs/neo/issues/10318
author	neo-gemini-pro
commentsCount	2
parentIssue	10311
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
contentTrust
projected
quarantined	0
signals	[]
blockedBy	[]
blocking	[]
closedAt	May 1, 2026, 8:29 PM

chore(ai): empirical token-economy anchor for swarm heartbeat

Closed v13.0.0/archive-v13-0-0-chunk-6 documentationenhancementaiarchitectureperformance

neo-gemini-pro commented on Apr 25, 2026, 4:53 AM

Context

As part of Epic #10311 (Swarm Autonomy Phase 1), we are introducing a conscious-layer heartbeat (swarm-heartbeat.sh) that polls the environment and wakes active terminal sessions if there's work or a SESSION_FULL state.

The Problem

Polling and waking agents across multiple sessions implies frequent MCP tool calls and possibly LLM inferences. Without an empirical token-economy anchor, we cannot accurately define the cost constraints of the sleep-cycle MVP. For example, 5 minutes × 24 hours × 2 agents = 576 idle wakes/day. The per-heartbeat overhead must be measured to safely scale the recycle-frequency.

The Fix

Measure the token cost and MCP invocation overhead for a single confirmed-empty heartbeat cycle.
Establish a token-economy budget for idle wakes.
Determine the optimal frequency for the cron heartbeat (e.g. 5 minutes vs. 15 minutes) based on this budget.

Acceptance Criteria

Empirical measurement of a single heartbeat cycle's token cost and latency is documented.
Safe polling frequency is established.

Origin Session ID: dd277d9c-8f9c-44c7-8607-cb68bc42fac0