What is the 'Neural Link'?

The Neural Link is a bi-directional bridge that connects AI agents (like Gemini or Claude) directly to the Neo.mjs runtime. It allows agents to 'see' the application's Scene Graph, inspect component state, verify event listeners, and even mutate the running application in real-time. This turns the application into a 'glass box' for AI, enabling autonomous debugging and feature development.

Why is Neo.mjs called an 'Application Engine' instead of a framework?

Traditional frameworks are general-purpose libraries (like a Toyota) that help you organize code, but they compile away into ephemeral DOM nodes. Neo.mjs is a precision-engineered runtime (like an F1 car), similar to Unreal Engine for games. It maintains a persistent 'Scene Graph' of objects in a separate worker thread. These objects retain their identity, state, and relationships, allowing for advanced capabilities like multi-window orchestration, runtime permutation, and deep AI introspection that are impossible with 'melted plastic' DOM-based frameworks.

What is 'Context Engineering'?

Context Engineering is the practice of curating the environment and information flow for AI agents to maximize their autonomy. In Neo.mjs, this is implemented via three Model Context Protocol (MCP) servers: a Knowledge Base for semantic code understanding, a Memory Core for learning from past sessions, and a GitHub Workflow server for project management. This ecosystem allows agents to work as fully integrated members of the development team.

What is 'Object Permanence' in the context of Neo.mjs?

In Neo.mjs, UI components are persistent JavaScript objects living in the App Worker, not just transient rendering results. This 'Object Permanence' means a component (like a Dashboard) maintains its state (scroll position, user input, internal logic) even if it is detached from the DOM or moved to a different browser window. This is the 'Lego Technic' approach versus the 'Duplo' approach of traditional frameworks.

What is an 'Agent Operating System'?

Neo.mjs v11 introduced the concept of an Agent OS, where the platform itself provides the tools and interfaces for AI agents to operate. It combines a standalone, type-safe AI SDK for autonomous 'Code Execution' with the Neural Link for runtime control. This enables agents to monitor, debug, and heal the application autonomously, effectively acting as an operating system for synthetic intelligence.

How does Neo.mjs handle multi-window applications?

Neo.mjs uses shared web workers to run a single application instance across multiple browser windows. All windows share the same application state and data in real-time. Components can even move between windows while retaining their JavaScript instances. This enables desktop-class experiences like multi-window IDEs, browser-based email clients, and multi-screen control rooms.

What makes Neo.mjs different from React, Angular, or Vue?

Neo.mjs is fundamentally different in its architecture: (1) It uses True Multithreading via web workers to prevent UI jank, (2) It is an AI-Native Application Engine with built-in bridges for AI collaboration, and (3) It treats components as persistent objects in a Scene Graph, enabling native multi-window support and runtime permutation.

Backlog
v12.1.0

9586 Analysis: Robust Multi-Window LivePreview Routing & Base Path Resolution
9585 Bug: LivePreview popout windowUrl generates incorrectly in dist environments
9584 Bug: Data worker broadly bundles non-data framework classes (Main.mjs) causing Webpack build failure
9582 Bug: Webpack fails due to unanchored `webpackInclude` picking up Node modules
9581 Bug: Webpack fails to resolve Task, Canvas, and Data worker dynamic chunks due to strict magic comments
9583 Bug: Webpack matches absolute paths, invalidating the ^\.\/ magic comment anchor
9580 Bug: Webpack fails to resolve internal app.mjs chunks due to strict magic comment regex
9579 Fix ChromaDB defrag crash on ghost entries
9578 Portal: Add TreeGrid Big Data Example
9577 Draft Release Notes: Storytelling & Narrative Arc
9575 Draft Release Notes: Gather Visual Assets
9576 Draft Release Notes: MCP Infrastructure & Security
9569 Epic: Neo.mjs v12.1 Release Notes
9574 Draft Release Notes: Generate Full Changelog
9573 Draft Release Notes: Learning Guides & Minor Features
9572 Draft Release Notes: ScrollSync & Performance
9571 Draft Release Notes: Locked Columns & Value Banding
9570 Draft Release Notes: TreeGrid & Data Pipeline Architecture
9568 Fix code examples in DataPipelines.md guide
9567 Update AI_QUICK_START.md to include Antigravity configuration
9539 Add TreeStore override for updateKey()
9566 Enhance `agent` parameter formatting in GitHub Workflow MCP Server
9538 Enhance Collection.updateKey() to support filtered collections (allItems map)
9565 Regression: Store.load({url}) fails when Store has no initial URL config
9552 Docs: Create Learning Guide for the Unified Data Pipeline Architecture
9564 Finalize Data Pipeline Push Integration & UI Reactivity
9563 Modularize MCP Server Architecture and Extract Shared Services
9562 Implement Generic OIDC Discovery and Google OAuth Hands-on Guide
9561 Add Functional Tests for MCP Authorization and Refine Error Handling
9560 Create Comprehensive Guide for MCP Server Authorization
9558 Implement OIDC/OAuth Authorization for KB and Memory MCP Servers
9551 Examples: Implement unified Data Pipeline showcases
9550 Refactor(data): Implement Store-to-Pipeline Legacy Bridge
9502 Migrate existing Stores to the new Pipeline architecture
9549 Remove `console.log` statements causing noise in unit tests
9548 `calcValueBands` broken state initialization on `splice` (StoreValueBanding test failure)
9454 Implement Push-Based WebSocket Integration in Data Pipeline
9455 Integrate RPC API into Pipeline Architecture (Connection.Rpc)
9547 Fix RemoteMethodAccess for Main Thread Addons and Instance-to-Instance ID collision
9546 Refactor Pipeline IPC to use Declarative Remote Configs (Instance Proxies)
9453 Implement Pipeline IPC and Remote Execution Routing
9545 Fix Knowledge Base Tool Argument Routing (x-pass-as-object)
9544 Enhance RemoteMethodAccess to Support Instance-to-Instance Routing
9543 Store Pipeline Instantiation and Legacy Parser Compatibility
9452 Connection Foundation and Parser Refactoring
9456 Epic: Buffered Grid - High-Performance Locked Columns (`locked: 'start' | 'end'`)
9542 Update Code of Conduct to include AI Entities & Autonomous Agents
9541 Remove outdated and redundant .github/CONCEPT.md
9540 Update CONTRIBUTING.md to include AI Agents & Autonomous Entities
9196 Feature: Add Collection.updateKey() method
9534 Rename mcp-stdio.mjs to mcp-server.mjs
9533 Migrate MCP servers from SSEServerTransport to StreamableHTTPServerTransport
9532 Support SSE transport for knowledge-base and memory-core MCP servers
9531 Grid SortZone incorrectly intercepts column resize drag events, causing VDOM corruption and layout thrash
9530 Grid cells vanish during column resize after a drag-and-drop column reorder
9529 Add lightweight cell resizing method to grid.Body
9528 Real-time cell resizing during grid column drag:move
9527 Grid cells do not shrink on column resize drop due to minWidth constraint
9526 Add visual focus styling during grid column resizing
9525 Fix header resize proxy visibility and prevent unwanted absolute styles on drop
9524 Prevent column sort toggle when clicking/dragging resize handles
9523 Resolve drag&drop collision between grid column resorting and resizing
9522 Grid Column Resizing: Header Logic and Update on Drop
9521 Tree Grid: Optional Helper Lines for Tree Structures
9520 LivePreview: Propagate active theme to popout child window
9519 SharedWorker: Fix theme resolution for child windows
9518 Fix LivePreview incorrect popout URL for subsequent windows (SharedWorkers)
9517 Workers: Standardize and optimize dynamic imports
7918 Epic: Neo Command Center (Agent OS UI)
8022 Refactor AgentOS Viewport: Modularize Components
8025 Create Custom Theme: Neo Cyberpunk
9516 Portal HomeCanvas: Enhance spark contrast in dark theme
9515 [Epic Sub] Value Banding styling conflicts with row selection
9514 [Epic Sub] TreeStore Value Banding Support
9513 Optimize Grid Value Banding (startIndex Support)
9512 Dynamic Value Banding Updates
9511 [Epic] Grid Value Banding
9510 Tooltip regression: dynamic domListeners bypassed reactivity and bubbling fixes
9509 Docs: Align core engine guides to use "engine" terminology instead of "framework"
9508 Docs: Explain Main Thread Addons and Lazy Loading in Lifecycle guide
9507 Docs: Clarify Two-Tier Reactivity Architecture and core.Config role
9506 Improve inline code styling within h1 tags in Markdown component
9505 Portal: Missing scrollbar contrast in learning section markdown component
9504 List: Fix vertical scrollbar visibility and contrast
9451 Create Pipeline Cornerstone and Refactor Store Implementation
8019 Bug: isWindowDragging flag stuck after external drop
9503 TimelineCanvas regression: Canvas not showing due to SharedCanvas getCanvasId() mismatch with componentId
9500 Epic: The Core Engine Learning Guides & Workshop
9485 Grid: Horizontal Scroll Performance & Jitter for Locked Columns
9484 Grid: Add Unit Tests for Locked Columns Feature
9460 Grid: Column Drag & Drop Integration & State Transitions
9459 Grid: Implement Reactive locked Config and Cell Pooling Bypass
9458 Grid: Create Main Thread Addon for Column Pinning (CSS Variables)
9483 Grid: Implement Reactive locked Config and Run-Time Column Reordering
9457 Grid: Implement Mathematical Column Layout Engine
9356 Chrome Windows: Multi Window Drag & Drop
9359 Google Windows Covid 19 Helix
9365 Chrome Window Neo mjs home
9482 Firefox Nightly: Scroll wheel stuck on Portal home page due to scroll-snap-type mandatory
9481 Auto-detect `isTreeGrid` config based on the Store instance
9479 Firefox Nightly R&D: OffscreenCanvas Worker-to-Worker Transfer Failure
9478 Grid ScrollManager: Support dynamic windowId changes for Main Thread Addon registration
9363 Chrome Window Learn funktionale Komponenten
9364 Chrome Window Calendar Basic
7957 Epic: WebSocket RMA Infrastructure
7958 Epic: Telemetry & Observability
7959 Epic: Agent Security & Capabilities
9476 DevIndex: Add Contextual Rank Column to Grid
9475 Grid: Remove `#initialChunkSize` DOM pooling bypass & fix pool shrink bug
9474 Grid: Optimize row pooling for small datasets
9473 Add File Editing Tool Protocol to AGENTS.md
9472 Add Testing Protocol to AGENTS.md
9471 Add Anti-Corruption Protocol to AGENTS.md
9470 Fix Grid Drag Scrolling Regression
7914 Epic: Foundation for Ticket-Driven Agent Orchestration
7919 Epic: Decouple AI Tooling for Public Ecosystem
7922 Feat: 'Derailment' Intervention Interface
7923 Refactor: Extract Memory Core to @neomjs/ai-memory-server
7924 Refactor: Extract GitHub Workflow to @neomjs/ai-github-server
7925 Feat: Create VisualService (Sighted Agent SDK)
7926 Feat: Add Cross-Repo Capabilities to GitHub Workflow MCP
9461 Epic: TreeGrid Big Data Demo
9469 TreeStore: internalIdMap desyncs after bulk operations (expandAll), breaking selection
9468 Fix memory leak in TreeStore.#rebuildKeysAndCount()
9466 TreeGrid Big Data Demo: Create E2E Tests
9467 Fix TreeStore.collapse() failing after expandAll() due to out-of-sync Collection map
9465 TreeGrid Big Data Demo: Styling, Logging, and Final Integration
9464 TreeGrid Big Data Demo: Implement Controls and Grid Wiring
9463 TreeGrid Big Data Demo: Implement Organic Data Generation Algorithm
9462 TreeGrid Big Data Demo: Scaffold Base Directory and Files
9420 Migrate Data Pipeline to Connection -> Parser -> Normalizer flow
9450 Enhance Data Worker to Instantiate Dynamically Loaded Modules
9419 Implement Dynamic Module Loading in `Neo.worker.Data`
9418 Create Data Normalizer Architecture (`Neo.data.normalizer.Base` & `Tree`)
9412 TreeGrid Component Tests (UI & Interactions)
9448 TreeGrid Component pooling accumulates `is-leaf` class leading to visual bugs
9447 TreeGrid: Fix 7-click expand/collapse bug and redundant change events
9446 Fix ComboBox test failure due to internalId changes
9445 TreeGrid Documentation: Create high-level architectural guide for TreeStore
9444 Clean up redundant `Neo.isDestroyed` catch blocks
9443 Stabilize Playwright Unit Tests by Centralizing Global Mocks
9442 Fix Grid Pooling DOM thrashing and Collection internalId lookup
9440 AI Workflow: Integrate Anchor & Echo strategy into Pre-Flight Commit check
9439 TreeStore: Apply "Anchor & Echo" JSDoc strategy for AI discoverability
9438 TreeStore: Reduce GC pressure and redundant iterations
9437 TreeStore: Optimize #allRecordsMap iteration loops
9433 TreeStore: Implement bulk expandAll() and collapseAll() methods
9435 TreeStore: Fix visible projection for dynamic child additions to expanded parents
9434 TreeStore: Decouple clearFilters() from legacy allItems pattern
9432 TreeStore: Override clear() to prevent memory leaks and split-brain states
9431 TreeStore: Fix ARIA desync (siblingIndex and siblingCount) after sort and filter
9429 TreeStore: Implement ancestor-aware filtering for filter override
9428 TreeStore: Implement hierarchical sorting for doSort override
9430 TreeModel: Introduce childCount to decouple isLeaf state from emptiness
9411 TreeGrid Unit Tests (Data & Logic)
9427 Fix TreeStore projection mutations and Row VDOM recycling
9426 Refactor Tree column and component reactivity
9425 Refactor Tree cell component architecture for performance
9424 Optimize `cellPoolSize` calculation in `grid.Body` to prevent DOM waste
9410 TreeGrid Documentation & Examples
9423 TreeStore Full CRUD Support & Structural Mutations
9422 Update TreeStore to manage ARIA sibling fields on mutations
9408 Grid Core Integration & TreeGrid Accessibility
9407 Create `Neo.grid.column.Tree` & Cell Component
9417 Optimize `TreeStore` Hot Paths (Performance)
9416 Add Error State & Events for Async Tree Loading
9413 Create Async Subtree Loading for `Neo.data.TreeStore`
9415 Support "Turbo Mode" in `Neo.data.TreeStore`
9414 Refactor `Neo.data.Store` to unify Record Hydration
9406 Create `Neo.data.TreeStore`
9405 Create `Neo.data.TreeModel`
9403 Performance: Eliminate Path scrollTop Layout Thrashing in getTargetData
9402 Performance: Eliminate Main Thread Layout Thrashing in getTargetData
9401 Performance: Proxy Getter Hoisting in Grid Row & Array Slice Optimization
9400 Performance: O(1) Fast Paths for onScrollCapture getById Traversal
9398 Performance: Eliminate O(N²) App Worker VDOM Traversal (syncVnodeTree)
9396 E2E: Implement Synthetic Thumb Drag Profiles for Optical Pinning Validation
9395 E2E: Refine GridRowScrollPinning to Target Explicit Thumb Drags
9394 E2E: Validate GridRowScrollPinning against DevIndex Canvas Worker Latency
9393 E2E: Implement GridRowScrollPinning via CSS Variables
9392 E2E: Implement GridRowScrollPinning Automated Test and Sync Refinement
9391 E2E: Refactor GridRowScrollPinning to Hybrid rAF Engine
9390 E2E: Fix GridRowScrollPinning Registration and DOM Lookup Flaws
9389 E2E: Remove Legacy Scroll Prediction Heuristics from Grid ScrollManager
9387 E2E: Implement Main Thread Optical Pinning for Grid Scrolling
9388 E2E: Enhance VDOM Update Pipeline with Meta Payload Support
7834 [Test] Button click event not firing on mobile
9386 E2E: Add min velocity threshold to Grid Predictive Scrolling
9385 E2E: Expose Performance tracker metrics via Remote Methods
9383 E2E: Implement Predictive Delta Injection (Velocity & Acceleration)
9382 E2E: Implement Dynamic RTT Measurement for VDOM Updates
9384 Implement generic Neo.util.Performance tracker
9381 E2E: Create Deterministic Grid Thumb Drag Benchmark
9379 Mobile Scrollbar UX: Increase touch width to 40px and theme thumb colors
9378 Extract desktop grid scrollbar width into a CSS variable
9377 Define `--grid-scrollbar-touch-width` variable across all themes
9369 [Epic] Upgrade ScrollSync for Loop-Free Two-Way Binding (Mobile & Desktop)
9376 Add .neo-has-touch capability class and enhance Mobile Grid Scrollbar
9374 Mobile UX Enhancements for Grid VerticalScrollbar (SCSS)
9373 Enable Mobile Two-Way ScrollSync for Grid VerticalScrollbar
9372 Refactor GridDragScroll to use ScrollSync API
9375 Improve ScrollSync Lock Release Mechanism (rAF vs setTimeout)
9371 API for Programmatic Scrolling in ScrollSync
9370 Upgrade ScrollSync Locking Mechanism
9361 Chrome Windows Johannes DevIndex
9362 Chrome Windows DevIndex
9360 scroll to bottom bug on main page
9368 Enhance Helix to rotate to initial selection on mount
9367 Fix Helix & Gallery selection models keynav and record resolution
9357 Chrome Windows Covid App
9358 Google Windows Covid 19 Helix
7847 Enhance WindowPosition Addon with Adaptive Tracking
9349 build: Fix dirent.path resolution and OS-agnostic pathing in esmodules.mjs
7756 Enhance `create_comment` tool to enforce reading related tickets

v12.0.0
v11.24.0
v11.23.1
v11.23.0
v11.22.0
v11.21.0
v11.20.0
v11.19.1
v11.19.0
v11.18.0
v11.17.1
v11.17.0
v11.16.0
v11.15.0
v11.14.0
v11.13.0
v11.12.0
v11.11.0
v11.10.0
v11.9.0
v11.8.0
v11.7.0
v11.6.1
v11.6.0
v11.5.0
v11.4.0
v11.3.0
v11.2.0
v11.1.0
v11.0.1
v11.0.0

7731 Fix: incorrect length check for releases object in ReleaseSyncer
7727 component.MagicMoveText: onVisibilityChange()
7726 main.DomEvents: add support for document.visibilitychange
7725 Docs: Update Memory Core API Guide for MCP SDK
7724 Docs: Update Agent-Agnostic MCP Config Guide for SDK Architecture
7723 Deprecate and remove `.github/mcp-servers.json`
7311 Fix MagicMoveText component cache issue causing jumping scroll
7713 health(github-workflow): reproduce gh-absent
7719 Docs: Clarify automatic session summarization in AGENTS_STARTUP.md
7718 Architectural Fix: Implement 'FIRST ACTION PROTOCOL' to ensure reliable session initialization
7716 Improve `create_issue` tool description for better user workflow
7715 Create .npmignore file to exclude tickets and release notes from npm package
7712 Investigate and Fix Race Condition Where Early Component Updates are Lost
7711 Fix VDOM Lifecycle and Update Collision Logic
7710 Refactor `vnodeInitialized` to a Reactive Config in `VdomLifecycle`
7709 Fix race conditions in component rendering and updates
7708 fix(ComboBox): Correct forceSelection behavior and update Playwright tests
7707 Major Performance & Modernization Refactor of JSDoc Documentation Generation
7706 Refactor jsdoc-x integration for improved performance and maintainability
7705 Remove Siesta from the project
7442 Convert Component Test form/field/ComboBox.mjs to Playwright
7443 Convert Component Test form/field/Text.mjs to Playwright
7441 Convert Component Test component/DateSelector.mjs to Playwright
7440 Convert Component Test component/Base.mjs to Playwright
7704 Enhance `worker.App` RMA methods to delegate worker errors
7703 Convert Component Test list/Chip.mjs to Playwright
7262 Enhance Development Workflow with Mandatory Unit Testing
7271 Convert Rectangle.mjs Test from Siesta to Playwright
7272 Convert VdomCalendar.mjs Test from Siesta to Playwright
7702 docs(agent): Improve session initialization protocol reliability
7701 docs: Strengthen agent protocols for tickets and memory
7700 docs: Correct Data Layer description in CodebaseOverview.md
7699 AI: Strengthen the mandatory pre-response check
7698 refactor(ai): Improve Separation of Concerns for memory-core server startup
7697 feat(ai): Automate knowledge base embedding on server startup
7696 Update AI_QUICK_START.md for new MCP server architecture
7695 Refactor: Split Agent Instructions for Persistent Enforcement
7694 Feat: Implement Automatic Agent Initialization via GEMINI.md
7693 Docs: Clarify Scope of Ticket-First Gate in AGENTS.md
7692 Docs: Clarify First Turn Memory Protocol in AGENTS.md
7691 Refactor: Harden Memory Core DB services against undefined `where` clauses
7690 Enhancement: Clarify Memory Core healthcheck response for hybrid database strategy
7689 Bug: Memory Core `querySummaries` fails when category is undefined
7405 MCP Server: Implement Knowledge Query Endpoint
7444 Create Component Test form/field/Number.mjs in Playwright
7634 docs(AGENTS.md): Revise Core Principles & Initialization
7296 Hacktoberfest 2025: Build Your AI Development Skills with Neo.mjs
7474 Improve AI Query Scoring to Boost Content Matches
7465 Hacktoberfest Mission: Implement the Memory Protocol Fix
7688 Integrate GitHub Workflow server tests into Playwright suite
7663 Feat: Implement GitHub API-based 'list_issues' Tool
7686 Enhance `button.Base` handler config for proper reactivity with closures
7685 data.Store lazy record conversion does not update allItems collection
7684 Fix: Memory Core Healthcheck `summarizationDetails` schema validation
7683 Dynamic two-way scroll sync for hybrid devices
7682 Implement unregister in ScrollSync addon
7681 Grid scrollbar: Improve two-way sync detection for hybrid devices
7680 Calendar regression: raw data access in store without creating the record
7679 ChromaManager.mjs: getMemoryCollection should use getOrCreateCollection
7476 Create Test Plan for form.field.Password
7439 Convert Component Test button/Base.mjs to Playwright
7675 Refactor: Standardize MCP server error response handling
7674 Fix(toolService): Create union type for multiple response schemas
7673 Refactor: Standardize score weights in Knowledge Base config
7062 functional.component.Base: enhance the `destroy()` signature
7671 Refactor: Polish Agent Protocols in AGENTS.md
7672 Refactor: Refine Query Tool Documentation in openapi.yaml
7668 Epic: Make Knowledge Base Tools Self-Documenting
7670 Refactor: Shorten AGENTS.md by removing query tool documentation
7669 Refactor: Enhance OpenAPI spec for Knowledge Base Server
7048 Epic: Refactor and Enhance the Email App
7666 Docs: Update AGENTS.md to use new Codebase Overview Guide
7604 Epic: Automate MCP Server Startup and Reduce Agent Protocol
7664 Docs: Create Codebase Overview Guide
7659 Docs: Create Self-Explanatory Tools via OpenAPI Descriptions
7662 Docs: Add Communication Style Directive to AGENTS.md
7630 docs: Update AGENTS.md to Reflect MCP Server Architecture
7661 Refine Memory Protocol: Clarify Recovery and Prevent Skipped Turns
7660 Refine Memory Protocol: Define a "Turn" and Consolidate Sub-Turns
7658 Refactor: Simplify Memory Core Protocol in AGENTS.md
7657 Docs: Add guide for troubleshooting tool calls
7656 Auto-start ChromaDB on MCP Server Startup - Implementation Summary
7655 Feat: Enhance Knowledge Base MCP Server Startup with Health Checks
7654 Feat: Enhance Knowledge Base Health Service and Update OpenAPI Schema
7653 Chore: Update MCP Server Names in `.gemini/settings.json` for Branding Consistency
7652 Feat: Re-enable `npm run` scripts for MCP Servers and Validate `npx` Readiness
7651 Fix: `IssueSyncer` fails to find releases after `ReleaseSyncer` refactoring
7649 Feat: Make `agent` parameter required for `create_comment` tool and simplify implementation
7648 Feat: Improve Agent Comment Formatting with Markdown Headings
7647 Fix: MCP Server Misinterprets Tool Error Responses
7646 Refactor: Streamline Release Metadata Handling in Sync Service
7643 Refactor: Implement MetadataManager for Sync Service
7642 Refactor: Extract Release & Issue Syncers from SyncService
7641 Refactor: Improve `createComment` with icon map and better formatting
7640 Feat: Add agent identity and templating to `createComment` tool
7639 Chore: Improve `create_issue` tool to enforce proposal workflow
7638 Feat: Add startup summarization status to healthcheck
7637 Fix: Race condition in async service initialization
7596 Bug: ChromaDB Fails on x64 Systems Without WSL
7633 bug(mcp): list_labels tool fails with schema validation error
7636 Refactor: Implement Robust Startup Sequence and Health Checks in Memory Core
7635 Refactor: Enhance Memory Core HealthService for Robustness and Diagnostics
7631 refactor: Remove obsolete buildScripts/ai directory
7204 Phase 3: Dynamic Proxy Transitioning (Windowing)
6996 Encourage Pure VDOM Effects
7264 Enhance Development Workflow with a Planning Phase
7306 Exploration Quest: Make the AI Aware of `buildScripts`
7316 AI Knowledge Evolution
7321 Create Guide for Personalized Agent Identity
7342 Hacktoberfest Mission: Train the AI - Test Its Memory! (1/5)
7343 Hacktoberfest Mission: Train the AI - Test Its Memory! (2/5)
7344 Hacktoberfest Mission: Train the AI - Test Its Memory! (3/5)
7364 Integrate GitHub CLI to Streamline Contribution Workflow
7373 Define a Clear Pull Request Review Protocol for the Agent
7371 Enable Agent to Diff a Pull Request
7378 Enhance Knowledge Base to Include GitHub Comments
7376 Define Hybrid API vs. Knowledge Base Workflow for Agent
7369 Update Agent Workflow to use Automated GitHub Issue Creation
7564 Epic: Implement Two-Way GitHub Synchronization for Issues
7568 Refine Issue Creation to be API-First
7391 Refactor createGhIssue.mjs to use fs-extra for consistency
7399 Architect AI Tooling as a Model Context Protocol (MCP) Servers
7401 Design Knowledge Base MCP Server API
7404 MCP Server: Implement Knowledge Read Endpoints
7406 MCP Server: Implement Knowledge Admin Endpoints
7410 MCP Server: Implement Memory Create Endpoint
7403 MCP Server: Implement Knowledge Server Scaffold
7411 MCP Server: Implement Memory Lifecycle Endpoints
7412 MCP Server: Implement Memory Admin Endpoints
7629 Implement `unassign_issue` Tool to Remove Specific Contributors
7628 Enhance `create_issue` Tool to Support Assignees
7625 Implement Tool to Assign Issues to Contributors
7627 Gemini CLI Test Issue for Unassignment
7626 Cache Viewer Permission on Server Startup
7590 Epic: Migrate GitHub MCP Services from GH CLI to GraphQL API
7595 Finalize GraphQL Migration & Cleanup
7624 Create Generic `get_viewer_permission` Tool
7623 Enhance GitHub Workflow OpenAPI Spec with Rich Descriptions
7622 Create MCP Tool for GitHub Issue Creation
7621 Refactor IssueService to Externalize GraphQL Queries
7620 Optimize SyncService by Scoping Push and Hashing Releases
7619 Refactor Magic Number for Release Fetching Pagination
7618 Optimize SyncService Release Fetching
7616 Neo.ai.mcp.server.github-workflow.SyncService: Switch to Content Hash Tracking
7615 Optimize GitHub Issue Sync: Dynamic `syncStartDate` for Delta Updates
7613 Update Pull Request Template for Enforced Issue Linking
7612 Implement Command-Line Debug Flag Override for MCP Servers
7611 Decouple Logger by Duplicating into Each MCP Server
7610 Refactor `LocalFileService` to Remove `glob` Dependency
7609 New Tool: Get Local Issue by ID
7608 Bug: `get_pull_request_diff` Fails Schema Validation Due to Raw String Output
7607 Add Release Name as H1 Header in Synced Release Note Files
7606 Rename `sync_issues` Tool to `sync_all` for Clarity
7605 Add Configurable Prefix for Release Note Filenames
7603 Bug: Issue Archiving Logic Incorrectly Buckets Recently Closed Issues
7602 MCP Tool Handlers Lose `this` Context, Causing Private Field Errors
7600 Update OpenAPI Specification to Reflect GraphQL Data Models
7599 Externalize GraphQL Queries for Label and PullRequest Services
7598 Finalize MCP Server Config Refactoring
7594 Migrate SyncService to GraphQL
7593 Migrate PullRequestService to GraphQL
7592 Migrate Label & Issue Services to GraphQL
7591 Implement GraphQL Client & Auth Service
7589 Refactor Issue Filename Convention
7582 Externalize Magic Numbers and Strings in SyncService
7580 Add Frontmatter to Synchronized Release Notes
7579 Track and Handle Failed Pushes in SyncService
7581 Add Comprehensive JSDoc to SyncService
7588 Enhance Sync Result Payload with Comprehensive Statistics
7578 Harden Release Caching and Add Fallback
7587 Implement Graceful Health Handling with Recovery
7576 Fix and Verify GitHub API Field Names in SyncService
7585 Implement Missing #checkGhAuth Method in HealthService
7584 Fix Semantic Version Comparison in HealthService using 'semver'
7583 Implement Abort-on-Startup if Health Check Fails
7577 Verify Minimum gh CLI Version Before Sync
7575 Implement Release Note Synchronization
7574 Update Archiving Logic to Use Dynamic Release Data
7573 Filter GitHub Issues by Date in SyncService
7572 Implement Dynamic Release Fetching in SyncService
7571 Refactor Sync Config for Dynamic Date-Based Syncing
7570 Implement and Adopt a Centralized stderr Logger
7569 Refactor SyncService to Use Centralized Configuration
7567 Implement Local to GitHub Sync (Push)
7566 Implement GitHub to Local Sync (Pull)
7565 Create Scaffold for the Issue Synchronization Service
7477 Architect GitHub Workflow as MCP Server
7536 Epic: Integrate Neo.mjs Core into MCP Servers
7563 Standardize ChromaDB Collection Naming Convention
7562 Refactor MCP Service ClassNames to Use Full Server Names
7561 Refactor AI Config for Server-Specific Namespacing
7560 Centralize GitHub Workflow Configuration
7559 Convert pullRequestService to PullRequestService Neo.mjs Class
7558 Convert labelService to LabelService Neo.mjs Class
7557 Convert issueService to IssueService Neo.mjs Class
7556 Convert GitHub Workflow healthService to HealthService Neo.mjs Class
7555 Centralize ChromaDB Client Warning Suppression in ChromaManager
7554 Convert queryService to QueryService Neo.mjs Class
7553 Refactor HealthService to Match Superior Memory Core Pattern
7552 Convert documentService to DocumentService Neo.mjs Class
7551 Convert databaseService to DatabaseService Neo.mjs Class
7550 Convert databaseLifecycleService to DatabaseLifecycleService Neo.mjs Class
7549 Create ChromaManager and Adjust KB Server Entry Point
7547 Enhance DatabaseLifecycleService with Eventing
7546 Convert textEmbeddingService to TextEmbeddingService Neo.mjs Class
7545 Convert summaryService to SummaryService Neo.mjs Class
7544 Convert sessionService to SessionService Neo.mjs Class
7543 Convert memoryService to MemoryService Neo.mjs Class
7542 Convert healthService to HealthService Neo.mjs Class
7540 Convert dbService to DatabaseService Neo.mjs Class
7539 Improve isDbRunning() in DatabaseLifecycleService
7538 Convert DatabaseLifecycleService to a Neo.mjs Class
7537 PoC: Refactor ChromaManager to a Neo.mjs Class
7535 Enhance MCP ToolService to Support OpenAPI Keywords
7534 Implement Centralized Logger for MCP Servers
7529 Epic: Implement Agent-Managed Database Tools
7531 Add Database Management Tools to Memory Core Server
7532 Make start_database Hybrid-Aware
7530 Add Database Management Tools to Knowledge Base Server
7520 Epic: Migrate Memory Server to stdio-based MCP
7524 Remove Legacy Express Server from Memory Core
7528 Use Centralized AI Config in Memory Core Services
7527 Refactor Health Service to Remove Redundant Try/Catch
7526 Refactor dbService to use chromaManager
7525 Enhance Memory Core Health Check
7523 Implement Memory Core toolService
7522 Refactor Memory Core OpenAPI for MCP
7521 Scaffold Memory Core MCP Server
7501 Architect AI Knowledge Base as MCP Server
7519 Dynamically determine argument passing strategy from OpenAPI spec
7518 Make toolService resilient to server-prefixed tool names
7517 Refactor toolService.mjs to Reduce Code Duplication
7516 Bug: query_documents input parameter mismatch
7515 Healthcheck tool returns unstructured content
7514 Enhance All Tools with In-Spec Manuals
7513 Enhance OpenAPI Spec with Tool Usage Examples
7512 Review and Correct Database Service Implementation
7511 Separate Create and Embed Services
7510 Refactor MCP Server Configuration
7509 Implement Document Retrieval Services
7508 Implement Sync Database Service
7507 Implement Query Documents Service
7506 Implement Delete Database Service
7505 Implement Healthcheck Service
7504 Implement Knowledge Base Tool Service
7503 Scaffold Knowledge Base MCP Server
7502 Design Knowledge Base MCP Server API
7500 Simplify Description Handling in Zod Schema Generation
7499 Implement Conditional Omission of Optional Fields in MCP Tool Definitions
7498 Fix Gemini CLI Client Compatibility for MCP `tools/list` Response
7497 Refine `mcp-stdio.mjs` for MCP Compliance and Clarity
7496 Add Annotations to MCP Tool Definitions
7495 Add outputSchema to MCP Tool Definitions
7494 Implement Zod-based Validation with JSON Schema Conversion
7493 Implement Dynamic Tool Discovery via OpenAPI
7492 Refactor to Direct MCP Tool Definitions
7491 Implement MCP Stdio Server for GitHub Workflow
7487 Enhance Tools List with OpenAPI Schema
7486 Refactor to an MCP Tool-Providing Server
7485 Manage Repository Labels
7484 Get PR Conversation History
7483 Implement PR Commenting Endpoint
7482 Enhance PR Listing and Checkout Logic
7481 Implement Pull Request API Endpoints
7480 Refine Health Check for GitHub Workflow Server
7479 Scaffold GitHub Workflow MCP Server
7478 Define Initial OpenAPI Specification for GitHub Workflow Server
7372 Enable Agent to Comment on Pull Requests
7381 Create a Robust GitHub CLI Setup and Verification Script
7366 Enable Agent to List GitHub Issues and PRs
7365 Document and Configure GitHub CLI Authentication
7464 MCP Server: Refine Memory Server Endpoints
7473 REFACTOR: Assign `type: 'blog'` directly in `createKnowledgeBase.mjs`
7467 `ai:query` fails to find content in existing blog post
7471 Refactor: Implement Granular Playwright Test Configurations
7470 PoC: Create Component Test for component.Image
7469 Refactor: Consolidate and Refine Playwright Test Harness
7437 Create Empty Viewport App for Component Tests
7468 MCP Server: Relocate Memory API Specification
7409 MCP Server: Implement Memory Query Endpoints
7408 MCP Server: Implement Memory Read Endpoints
7407 MCP Server: Implement Memory Server Scaffold
7436 Create Component Test Harness Config
7449 Create Sitemap & LLM.txt Generator Script
7438 Create RMA Helper Scripts for Component Tests
7445 Add `loadModule` RMA Method to App Worker
7434 Restructure BACKERS.md into an Investment Prospectus
7287 Convert state/FeedbackLoop.mjs Test from Siesta to Playwright
7418 Refine and Integrate GitHub CLI Guide
7432 Refine and Align Core Project Documentation (VISION.md & README.md)
7431 Refine Project VISION.md with Core Architectural Concepts
7265 Rewrite Project Vision (`VISION.md`)
7430 Update and Formalize Project VISION.md
7429 Refine and Update Project Roadmap
7428 Update and Formalize Project ROADMAP.md
7377 Create Script to Sync New GitHub Issues to Local Markdown Files
7282 Convert functional/Button.mjs Test from Siesta to Playwright
7417 Polish the new 'Getting Started' Guide
7426 Docs: Modernize CONTRIBUTING.md
7425 MCP Config: Align Knowledge Server Port and Health Check
7387 Create Guide for Chrome DevTools MCP Server
7283 Convert functional/HtmlTemplateComponent.mjs Test from Siesta to Playwright
7402 Design Memory Core MCP Server API
7297 Create a Modern 'Getting Started' Guide
7393 Move PR workflows to separate guide in learn/guides/ai
7270 Convert ManagerInstance.mjs Test from Siesta to Playwright
7400 Define Agent-Agnostic MCP Server Configuration
7286 Convert state/createHierarchicalDataProxy.mjs Test from Siesta to Playwright
7414 Implement GitHub Action to Prevent Issue Reopening
7413 MCP Server: Implement Knowledge Create/Update Endpoint
7398 Ensure Cross-Platform UUID Generation for Agent Memory
7397 Update AGENTS.md with Correct Health Check Endpoint
7394 Clarify UUID Generation for Agent Memory
7370 Enable Agent to Checkout PR Branches
7386 Configure MCP Server for Project
7346 Hacktoberfest Mission: Train the AI - Test Its Memory! (5/5)
7345 Hacktoberfest Mission: Train the AI - Test Its Memory! (4/5)
7368 Enhance Issue Creation Script to Sync GitHub ID back to Markdown File
7367 Create Script to Automate GitHub Issue Creation from Markdown Ticket
7363 Create `clearSummaries.mjs` script for development
7362 Correct Agent Initialization Workflow
7358 Refactor `summarizeSession.mjs` to automatically summarize all un-summarized sessions
7361 Clarify Agent Memory Server Port in AGENTS.md
7293 Convert vdom/layout/Cube.mjs Test from Siesta to Playwright
7284 Convert functional/Parse5Processor.mjs Test from Siesta to Playwright
7289 Convert state/ProviderNestedDataConfigs.mjs Test from Siesta to Playwright
7292 Convert vdom/VdomRealWorldUpdates.mjs Test from Siesta to Playwright
7291 Convert vdom/VdomAsymmetricUpdates.mjs Test from Siesta to Playwright
7288 Convert state/Provider.mjs Test from Siesta to Playwright
7290 Convert vdom/Advanced.mjs Test from Siesta to Playwright
7357 Create and Integrate `ai:get-last-session` Script
7356 Clarify ai:add-memory command in AGENTS.md
7341 Correct agent memory initialization protocol
7338 Implement Session Recovery Protocol in AGENTS.md
7337 Enhance Agent Session Initialization: Generate New Session ID and Validate Memory Core State
7336 Implement Automated Session Summarization Workflow
7335 Clarify Agent Memory Protocol and Tooling
7334 Document Human Verification of Agent Memory
7320 Update Agent Workflow for Memory
7333 Create AI Strategic Workflows Guide
7332 Enhance Session Summary with Rich Metadata
7294 Convert vdom/table/Container.mjs Test from Siesta to Playwright
7285 Convert neo/MixinStaticConfig.mjs Test from Siesta to Playwright
7329 Add missing @reactive tags and fix config JSDoc for trailing underscores
7327 Clarify git-ignored class-hierarchy.yaml in AGENTS.md
7326 Document Optional Memory Core Setup
7325 Create Session Summarization API
7319 Create Memory Query Tool
7322 Implement Memory Backup and Restore
7318 Create Memory Capture API
7324 Refactor and Centralize AI Configuration
7317 Set up Memory ChromaDB
7281 Convert form/field/AfterSetValueSequence.mjs Test from Siesta to Playwright
7315 collection.Base#filter() should use constructor for allItems
7280 Convert core/EffectBatching.mjs Test from Siesta to Playwright
7301 Add Intent-Driven JSDoc to `splitter.MainContainer` Example
7299 Add Intent-Driven JSDoc to Tab Container Example
7279 Convert config/MultiLevelHierarchy.mjs Test from Siesta to Playwright
7278 Convert config/MemoryLeak.mjs Test from Siesta to Playwright
7277 Convert config/Hierarchy.mjs Test from Siesta to Playwright
7312 Fix Prototype Pollution in Neo.setupClass
7276 Convert config/CustomFunctions.mjs Test from Siesta to Playwright
7275 Convert config/CircularDependencies.mjs Test from Siesta to Playwright
7274 Convert config/Basic.mjs Test from Siesta to Playwright
7273 Convert config/AfterSetConfig.mjs Test from Siesta to Playwright
7269 Convert CollectionBase.mjs Test from Siesta to Playwright
7310 Improve Community Channel Discoverability in README
7309 Refine tab.Container JSDoc and Update AGENT.md Example
7308 Create "Working with AI Agents" Guide
7303 Add Temporary Hacktoberfest Onboarding Protocol to AGENTS.md
7302 Add Intent-Driven JSDoc to `splitter.MainContainer` Example
7300 Refactor Knowledge Base Schema to Fix Type Regression
7295 Update AI Quick Start Guide
7268 Convert ClassSystem.mjs Test from Siesta to Playwright
7267 Convert ClassConfigsAndFields.mjs Test from Siesta to Playwright
7263 Plan Remaining Siesta to Playwright Test Migrations
7261 Convert core/Effect.mjs Siesta Test to Playwright
7260 Correct Playwright Output Directory
7258 Refactor Playwright Setup and Convert VdomHelper Test
7259 VdomHelper incorrectly moves className into attributes
7257 Refine AI Agent Guidelines Workflow
7256 Refactor Playwright Setup Function for Test-Specific Configurations
7255 Refactor Playwright Test Configuration
7254 Install Playwright and Convert First Siesta Test
7253 Update AI Agent Guidelines in AGENTS.md
7252 Add Ticket Archive to Knowledge Base
7251 Finalize and Document Ticket Archival Process
7250 Improve Agent Information Discovery Protocol

v10.9.0
v10.8.0
v10.7.0
v10.6.0
v10.5.4
v10.5.3
v10.5.2
v10.5.1
v10.5.0
v10.4.1
v10.4.0
v10.3.5
v10.3.4
v10.3.3
v10.3.2
v10.3.1
v10.3.0
v10.2.1
v10.2.0
v10.1.1
v10.1.0
v10.0.2
v10.0.1
v10.0.0
v10.0.0-beta.6
v10.0.0-beta.5
v10.0.0-beta.4
v10.0.0-beta.3
v10.0.0-beta.2
v10.0.0-beta.1
v10.0.0-alpha.5
v10.0.0-alpha.4
v10.0.0-alpha.3
v10.0.0-alpha.2
v10.0.0-alpha.1
v9.16.0
v9.15.0
v9.14.0
v9.13.1
v9.13.0
v9.12.0
v9.11.1
v9.11.0
v9.10.5
v9.10.4
v9.10.2
v9.10.1
v9.10.0
v9.9.0
v9.8.0
v9.7.1
v9.7.0
v9.6.1
v9.6.0
v9.5.0
v9.3.0
v9.2.0
v9.1.0
v9.0.2
v9.0.1
v9.0.0
v8.43.1
v8.43.0
v8.42.0
v8.41.2
v8.41.1
v8.41.0
v8.40.0
v8.39.1
v8.39.0
v8.38.0
v8.37.0
v8.36.0
v8.35.1
v8.35.0
v8.34.0
v8.33.0
v8.32.0
v8.31.1
v8.31.0
v8.30.0
v8.29.0
v8.28.1
v8.28.0
v8.27.1
v8.27.0
v8.26.1
v8.26.0
v8.25.0
v8.24.0
v8.23.0
v8.22.0
v8.21.2
v8.21.1
v8.21.0
v8.20.2
v8.20.1
v8.20.0
v8.19.1
v8.19.0
v8.18.1
v8.18.0
v8.17.1
v8.17.0
v8.16.0
v8.15.0
v8.14.3
v8.14.2
v8.14.1
v8.14.0
v8.13.0
v8.12.0
v8.11.0
v8.10.1
v8.10.0
v8.9.1
v8.9.0
v8.8.0
v8.7.1
v8.7.0
v8.6.2
v8.6.1
v8.6.0
v8.5.0
v8.4.2
v8.4.1
v8.4.0
v8.3.0
v8.2.0
v8.1.3
v8.1.2
v8.1.1
v8.1.0

Frontmatter

id	7695
title	Refactor: Split Agent Instructions for Persistent Enforcement
state	Closed
labels	enhancementairefactoring
assignees	tobiu
createdAt	Nov 2, 2025, 7:57 PM
updatedAt	Nov 2, 2025, 8:44 PM
githubUrl	https://github.com/neomjs/neo/issues/7695
author	tobiu
commentsCount	0
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	Nov 2, 2025, 8:44 PM

Refactor: Split Agent Instructions for Persistent Enforcement

Name: Neo.mjs Application Engine
Author: Neo.mjs

Closed v11.0.0 enhancementairefactoring

tobiu commented on Nov 2, 2025, 7:57 PM

This ticket tracks the refactoring of agent instructions to ensure persistent rule enforcement. The current AGENTS.md will be split into two files: AGENTS_STARTUP.md for one-time session initialization, and AGENTS.md for the per-turn enforcement rules. The .gemini/GEMINI.md trigger file will be updated to orchestrate this new flow, importing the new AGENTS.md to keep the rules in the context of every turn.

tobiu added the enhancement label on Nov 2, 2025, 7:57 PM

tobiu added the ai label on Nov 2, 2025, 7:57 PM

tobiu added the refactoring label on Nov 2, 2025, 7:57 PM

tobiu assigned to @tobiu on Nov 2, 2025, 7:58 PM

tobiu referenced in commit f7bf100 - "Refactor: Split Agent Instructions for Persistent Enforcement #7695" on Nov 2, 2025, 8:43 PM

tobiu closed this issue on Nov 2, 2025, 8:44 PM