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	10333
title	Implement since_comment_id and comment_id parameters for get_conversation
state	Closed
labels	enhancementaiarchitectureperformance
assignees	neo-gemini-pro
createdAt	Apr 25, 2026, 9:19 PM
updatedAt	Apr 25, 2026, 10:11 PM
githubUrl	https://github.com/neomjs/neo/issues/10333
author	neo-gemini-pro
commentsCount	2
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
contentTrust
projected
quarantined	0
signals	[]
blockedBy	[]
blocking	[]
closedAt	Apr 25, 2026, 10:11 PM

Implement since_comment_id and comment_id parameters for get_conversation

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

neo-gemini-pro commented on Apr 25, 2026, 9:19 PM

Context

#10272 highlighted that A2A PR reviews suffer from quadratic token scaling because get_conversation fetches the entire thread every time. We agreed exact-match fetching is needed to solve this.

The Problem

The read-path for PR conversations currently lacks selective fetch. Agents must parse the full history to find new comments, wasting context tokens.

The Architectural Reality

The get_conversation tool in ai/mcp/server/github-workflow/ uses the gh CLI. GitHub's REST API supports per-ID fetch, but our wrapper doesn't expose it.

The Fix

Add optional comment_id, since_comment_id, and last_n parameters to get_conversation with proper precedence (comment_id > since_comment_id > last_n > full).

Acceptance Criteria

get_conversation accepts comment_id, since_comment_id, last_n optional parameters.
Backward compatibility is maintained if no new parameters are passed.

Out of Scope

New tools like get_comment and list_comments are explicitly out of scope to respect the tool-surface cap.

Origin: #10272

Origin Session ID: 0b29a8fa-c6b0-42e2-ab3b-8015a99db2d8 Retrieval Hint: "selective fetch since_comment_id get_conversation a2a"