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.

Neo.mjs - Self-Evolving Software Organism

Frontmatter

title	fix(ai): bound topology REM logs (#13995)
author	neo-gpt
state	Merged
createdAt	Jun 25, 2026, 11:45 AM
updatedAt	Jun 25, 2026, 11:57 AM
closedAt	Jun 25, 2026, 11:56 AM
mergedAt	Jun 25, 2026, 11:56 AM
branches	`dev` ← `codex/13995-topology-rem-log-hygiene`
url	https://github.com/neomjs/neo/pull/13996
contentTrust
projected
quarantined	0
signals	[]

Merged

neo-gpt commented on Jun 25, 2026, 11:45 AM

Resolves #13995

Bounds REM topology extraction so it reserves graph output tokens, passes maxCompletionTokens to the provider, fails loud on context overflow, length truncation, and parse failure, and records bounded per-chunk and zero-conflict outcomes. Adds a durable file-only diagnostic channel and routes routine HealthService cache churn into it so the LMS developer console is not spammed while diagnostic logs are still retained.

Evidence: L2 unit/static covers topology request bounds, parse-failure friction, zero-conflict outcome logging, and file-only diagnostic behavior. L3 post-merge live LMS observation remains required for console-noise verification. Residual: post-merge LMS developer-console smoke check.

Deltas from ticket

#13984 remains closed; this PR targets the new #13995 only.

#13994 remains separate for the suspected 400k provider-context/counter contamination; this PR makes topology calls bounded and observable but does not claim to close that investigation.

Local ignored config overlays were refreshed with node ./ai/scripts/setup/initServerConfigs.mjs --migrate-config after rebasing onto origin/dev; no tracked config overlay changes.

Test Evidence

node --check ai/services/graph/TopologyInferenceEngine.mjs
node --check ai/services/memory-core/HealthService.mjs
node --check ai/mcp/server/shared/logger.mjs
node --check test/playwright/unit/ai/daemons/orchestrator/services/DreamService.spec.mjs
node --check test/playwright/unit/ai/services/memory-core/HealthService.spec.mjs
node --check test/playwright/unit/ai/mcp/server/shared/logger.spec.mjs
git diff --check origin/dev...HEAD
npm run test-unit -- test/playwright/unit/ai/daemons/orchestrator/services/DreamService.spec.mjs test/playwright/unit/ai/services/memory-core/HealthService.spec.mjs test/playwright/unit/ai/mcp/server/shared/logger.spec.mjs -> 106 passed

Post-Merge Validation

Run the live LMS developer console and diagnostic daemon path long enough to verify routine health cache churn no longer dominates user-facing logs while durable file logs still capture those diagnostics.

Commits

81486c7124 — fix(ai): bound topology REM logs (#13995)

Authored by Euclid (GPT-5, Codex Desktop). Session 1583113d-4deb-455a-bee4-16521e9d1479.

tobiu APPROVED reviewed on Jun 25, 2026, 11:56 AM

No review body provided.