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	10283
title	Extend Pre-Authoring Adjacency Sweep to Meta-Skills (create-skill)
state	Closed
labels	[]
assignees	[]
createdAt	Apr 24, 2026, 3:51 AM
updatedAt	Apr 24, 2026, 11:10 PM
githubUrl	https://github.com/neomjs/neo/issues/10283
author	neo-gemini-pro
commentsCount	0
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Apr 24, 2026, 11:10 PM

Extend Pre-Authoring Adjacency Sweep to Meta-Skills (create-skill)

Closed v13.0.0/archive-v13-0-0-chunk-6 []

neo-gemini-pro commented on Apr 24, 2026, 3:51 AM

Author's Note: Autonomously filed by Gemini 3.1 Pro based on retrospective insights from PR #10282.

Context

During the ideation-sandbox refactor (#10281 / #10282), we observed an architectural misstep: both agents missed the Progressive Disclosure mandate when initially modifying ideation-sandbox/SKILL.md. The gap was not a lack of knowledge, but a lack of invocation discipline. We filed and implemented skill enhancements without first consulting the authoritative skill-authoring guide (create-skill).

Proposal

Extend the "Pre-Authoring Adjacency Sweep" (introduced in #10276) from just content-level sweeps (checking prior ideation/tickets) to include a meta-level sweep.

Acceptance Criteria:

Update ticket-create/references/ticket-create-workflow.md (or relevant skill) to include a mandatory pre-flight check: "Before proposing/implementing skill changes, have you consulted .agent/skills/create-skill/SKILL.md to ensure adherence to Progressive Disclosure and other conventions?"
Add explicit instruction to the ticket-intake gate when the assigned ticket involves modifying any file in .agent/skills/.

This closes the exact failure mode that produced the bloat corrected in #10281.

tobiu referenced in commit 560b2c2 - "docs(ai): implement meta-skill adjacency sweep for skill modifications (#10283) (#10286) on Apr 24, 2026, 11:10 PM

tobiu closed this issue on Apr 24, 2026, 11:10 PM