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	12698
title	Mine and verify v13 release-note war stories
state	Closed
labels	documentationenhancementai
assignees	neo-gpt
createdAt	Jun 7, 2026, 9:47 PM
updatedAt	Jun 8, 2026, 2:52 AM
githubUrl	https://github.com/neomjs/neo/issues/12698
author	neo-gpt
commentsCount	8
parentIssue	12696
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Jun 8, 2026, 2:52 AM

Mine and verify v13 release-note war stories

Closed v13.0.0/archive-v13-0-0-chunk-16 documentationenhancementai

neo-gpt commented on Jun 7, 2026, 9:47 PM

Context

Parent epic #12696 defines v13 as a release epic, not a single release-note ticket. Its decomposition names war-story mining and verification as the first feed lane for the release-note MD spine, while #12695 remains the assembly artifact and PR #12697 carries the first ready-for-review draft.

This ticket exists because the v13 release-note bar is higher than an anchor list. Existing major release notes such as resources/content/release-notes/chunk-2/v12.0.0.md and resources/content/release-notes/chunk-2/v12.1.0.md are narrative, evidence-backed releases with memorable engineering stories. Parent epic #12696 explicitly asks for Symptom -> Investigation -> Culprit -> Fix arcs that feed the MD, not a raw changelog dump.

Pre-creation evidence on 2026-06-07:

Live latest-open sweep: checked the latest 20 open issues with number, title, author, labels, and URL. Found #12696, #12695, and other v13 backlog items, but no existing war-story mining leaf.
Exact live sweeps: v13 release war story mining, release notes war story, and v13 war stories returned only the parent epic or unrelated historical content.
Exact local sweep across resources/content/issues and resources/content/discussions for war-story, war story, release-note.*story, v13.*story, release notes.*verification, and Symptom.*Investigation.*Culprit.*Fix surfaced Discussion #12694 and unrelated historical references, but no equivalent active leaf ticket.
Semantic KB ticket sweep for ticket v13 release notes war story mining verification release artifact appendix did not identify an equivalent ticket.
Project pre-flight: Project 12 is live as Neo v13 Release; this ticket is attached to Project 12 for release visibility.

The Problem

The first v13 release-note draft now exists, but it still risks reading as a curated anchor list with one hero story rather than a v12-caliber engineering narrative. Parent epic #12696 already names a verified emblem story: #10370, where an emergent Claude Opus 4.7 Sunset Protocol became the session-sunset skill. It also names likely candidates such as B3 cluster-batch and ADR 0019 / AiConfig SSOT work.

Without a dedicated mining/verification leaf, every review cycle on #12697 would either ask the release-note author to rediscover stories ad hoc or push unrelated mining scope into the MD assembly PR. That blurs the boundary between source recovery and prose assembly.

The Architectural Reality

The release-note artifact is consumed by the release-note content path under resources/content/release-notes/, the generated release index in apps/portal/resources/data/releases.json, and later release-prep SEO surfaces. The war-story mining output is an input to #12695, not a replacement for it.

The required evidence can be recovered from public, repo-owned substrates:

GitHub issues and PRs, especially #10370 and PRs referenced by #12696.
Discussion #12694 and parent epic #12696.
Existing release-note precedent in resources/content/release-notes/chunk-2/v12.0.0.md and resources/content/release-notes/chunk-2/v12.1.0.md.
Memory Core only when healthy enough to provide corroborating history; the ticket must remain actionable via GitHub/git evidence if embeddings are degraded.

The Fix

Mine and verify the v13 engineering war stories that should feed the final release-note MD.

Expected output shape:

A curated set of release-note candidate stories, each with Symptom -> Investigation -> Culprit -> Fix structure.
Each story has public evidence anchors: issue numbers, PR numbers, discussion links, or committed path references.
Each story is classified by target section in #12695: Agent OS, Swarm Governance, Brain-on-Body, GitHub Workflow/Reliability, Body/Grid/UI, or External Providers/Deployment.
Weak or unverifiable stories are explicitly rejected rather than smuggled into public release prose.
The output is delivered as a comment or PR update that #12695 can consume without bloating #12695 into broad mining work.

Contract Ledger Matrix

Target Surface	Source of Authority	Proposed Behavior	Fallback	Docs	Evidence
War-story feed for #12695	Parent epic #12696 and Discussion #12694	Provide verified story candidates that strengthen the v13 release-note MD	If a candidate lacks public evidence, reject it or mark it as non-public/internal-only	#12695 release-note body / review response	Issue, PR, discussion, or committed path anchors
`Symptom -> Investigation -> Culprit -> Fix` shape	#12696 decomposition and v12.x release-note precedent	Normalize candidate stories into comparable engineering arcs	Keep as a short anchor-only mention if the full arc is not recoverable	Candidate-story comment or PR update	Public citation per story
Memory Core usage	Parent epic #12696 caveat about degraded mining	Use Memory Core only as corroboration, not as the sole public source	Fall back to GitHub/git evidence when embeddings are degraded	Mining notes	Memory query ID optional; public source required

Decision Record impact

None. This is release-content evidence mining; it does not amend an ADR.

Discussion Criteria Mapping

Source criterion	Ticket acceptance mapping
#12696 war-story bar	AC1-AC4 require verified `Symptom -> Investigation -> Culprit -> Fix` candidates.
#12694 appendix / no mega-changelog constraint	AC5 keeps mining curated and section-targeted, not a full PR dump.
#12696 public-surface guardrails	AC6 rejects weak, private, client-specific, or forward-commercial claims.

Acceptance Criteria

AC1: At least three candidate v13 war stories are mined and documented with Symptom -> Investigation -> Culprit -> Fix structure.
AC2: #10370 / session-sunset is verified as the hero Agent OS story and either accepted into #12695 or explicitly replaced with a stronger public candidate.
AC3: Each accepted candidate has public evidence anchors: GitHub issue, PR, Discussion, or committed path references. Memory Core may corroborate but is not the only evidence source.
AC4: Candidates are classified by release-note section so #12695 can consume them without restructuring the whole MD.
AC5: The output does not attempt to enumerate all 1,228 PRs or 1,606 closed issues; appendix generation remains a separate leaf.
AC6: Public-surface hygiene is checked: no client names, no private strategy, no unverifiable forward-looking claims.
AC7: The final output links back to #12695 / PR #12697 with concrete prose-update recommendations or a direct follow-up commit plan.

Out of Scope

Generating the exhaustive PR/issue appendices.
Editing the release-notes skill #10321.
Publishing the release or changing package versions.
Finalizing the Body/Grid section before the wrapper selection-model lane converges.
Creating per-section drafting tickets unless review of #12697 identifies a named evidence gap.

Avoided Traps

Rejected: pre-splitting #12695 into five section tickets before review. Parent epic #12696 has enough structure for one assembly PR plus feed leaves; section splits should be evidence-driven.
Rejected: private-memory-only stories. Public release notes require public evidence anchors.
Rejected: raw mega-changelog mining. Appendix generation is a separate release-blocking leaf.
Rejected: treating #10321 as this ticket. The skill remains post-release so it can learn from the artifact workflow.

Parent epic: #12696
Release-note MD spine: #12695
Current draft PR: #12697
Verified hero-story source: #10370
Release-note narrative Discussion: #12694
Future release-notes skill: #10321

Origin Session ID: e8f07ef9-ef7e-4815-8ff4-7abe13720621

Retrieval Hint: v13 release war story mining session-sunset 10370 ADR 0019 AiConfig B3 cluster batch release notes Symptom Investigation Culprit Fix

tobiu referenced in commit 9520a96 - "docs(release): add v13 release-note checkpoint (#12695) (#12697) on Jun 8, 2026, 12:32 AM

tobiu referenced in commit 8ee2f67 - "docs(release): update v13 Body/Grid cut-line (#12698) (#12714)" on Jun 8, 2026, 2:52 AM

tobiu closed this issue on Jun 8, 2026, 2:52 AM