NeoOptimize is an AI-empowered Windows optimization and maintenance platform for local diagnostics, safe cleanup, repair guidance, update integrity checks, real-time system monitoring, and before/after maintenance reporting.

The public release focuses on the NeoOptimize desktop application, local safety workflow, mini tray companion, update manager, NEO local assistant features, and release-gate validation.

Latest release-gate screenshot captured from the Windows VM validation run.

Verify the installer before running it:

Get-FileHash .\NeoOptimize.exe -Algorithm SHA256

The hash must match:

7b202f7dae8516f98ffbb6475e799432e1c4c1bb6a0ad9c7e63d323312b79493

The installer is also published as a GitHub Container Registry package for automation:

docker pull ghcr.io/neooptimize/neooptimize:1.0
docker run --rm ghcr.io/neooptimize/neooptimize:1.0 --sha256
docker run --rm -v "$PWD:/out" ghcr.io/neooptimize/neooptimize:1.0 --copy /out

NeoOptimize is built for practical Windows maintenance. It is not a cosmetic one-click optimizer. The application reads local system signals, explains probable causes, recommends safe actions, and keeps high-impact changes behind explicit user approval.

Core goals:

• Detect performance bottlenecks across CPU, GPU, RAM, disk, network, startup, services, updates, and power configuration.
• Explain system health in a format normal users and technicians can act on.
• Separate scan, report, cleanup, repair, benchmark, and update operations.
• Reduce accidental damage by ranking actions by risk and reversibility.
• Produce before/after reports so maintenance results can be measured.
• Keep local-first operation available without requiring a cloud account.

• CPU usage, kernel pressure, and clock information.
• GPU availability and usage when the local system exposes compatible counters.
• RAM usage, available memory, committed memory pressure, and cache indicators.
• Disk free space, read/write pressure, queue length, and latency indicators.
• Network throughput, adapter status, local IP information, and connectivity state.
• Uptime, process count, thread count, handle count, and power-state visibility.

AI Doctor turns local telemetry into a practical care plan. It prioritizes actions based on expected benefit, risk level, privilege requirement, and reversibility.

It can help explain:

• high idle CPU usage,
• RAM pressure,
• disk queue spikes,
• slow boot,
• network instability,
• Windows Update failures,
• repeated repair failures,
• unusual process growth,
• common security posture problems.

The mini tray is designed to stay lightweight while the main window is minimized. It gives quick access to monitoring and NEO interaction without opening extra PowerShell windows.

Mini tray capabilities:

• compact live CPU, RAM, and disk status,
• NEO text conversation panel,
• voice command entry point,
• clear chat action,
• quick-open button for the main interface,
• update and provider status shortcuts,
• automatic chat retention cleanup to avoid long local chat buildup.

NEO stands for Neural Execution Operator. It is the local assistant layer inside NeoOptimize.

When asked who it is, NEO answers:

I am NEO, Neural Execution Operator, an artificial intelligence built at Zenthralix-Lab by nol_eight.

NEO responsibilities:

• summarize local telemetry,
• explain anomalies,
• draft safe maintenance plans,
• guide the user through repair workflows,
• prepare PowerShell or CMD maintenance drafts for review,
• support text and voice interaction from the mini tray,
• keep actions tied to allowlisted NeoOptimize modules,
• require approval before executing privileged maintenance.

Local model support is designed for privacy-preserving offline diagnosis. The installer can download and install Ollama locally, start the local API hidden, and prepare neo:latest and neo-light:latest for NEO. Optional cloud providers can be configured by the user, but core local diagnostics remain available without requiring an API key.

The Update Manager follows a cautious, Linux Mint style workflow:

1. Check for update metadata.
2. Show release information and version details.
3. Verify SHA-256 before install.
4. Apply the update only after confirmation.
5. Repair the local installation if required files are missing or damaged.

Update actions are credential-gated and integrity-checked so an update package cannot be silently replaced without detection.

The 1.0 release gate completed with:

• Fixed-scale desktop UI validation at 1024x680 from the current Windows VM screenshot.
• Single-instance desktop guard validation for NeoOptimize startup.
• NEO Mini voice command label and tooltip validation.
• NEO Mini local AI fallback validation for status, anomaly scan, code repair guidance, corpus-aware suggestions, and Local AI setup prompts.
• Hidden Ollama setup validation: installer starts the Local AI helper in a background worker with no visible CMD/PowerShell window.
• Live endpoint smoke for /health, /healthz, /readyz, /livez, /api/v1/health, /api/v1/metrics, and /downloads/NeoOptimize.exe.
• Browser smoke across dashboard routes with no console errors and no HTTP 4xx/5xx responses.
• Server Jest suite: 69/69 tests across 21 suites.
• Python AI engine tests: 3/3.
• Client-nextgen production build, Rust/Tauri Windows build, public bundle verifier, static no-CMD-popup scan, and installer rebuild.

NeoOptimize is audit-first by default.

• Scan and report actions are separated from cleanup and repair actions.
• High-impact actions require administrator approval.
• Repair workflows are confirmation-gated and run behind the native UI through hidden workers.
• Cleanup tasks are scoped to known safe locations.
• Privileged execution still requires Windows Administrator approval when required.
• Remote access readiness features are disabled unless explicitly enabled by an administrator.
• Camera, microphone, browser secrets, private keys, documents, and biometric data are not collected by default.
• Reports stay local unless the user chooses to share them.

NeoOptimize is designed as a local-first maintenance utility. The public build does not require a user account for local diagnosis. Network access is used for release downloads, update metadata, optional support links, and optional user-configured AI providers.

1. Open the release page.
2. Download NeoOptimize.exe.
3. Verify the SHA-256 checksum.
4. Run the installer as Administrator.
5. Let the installer finish local AI setup. It downloads/installs Ollama if needed, starts ollama serve hidden, and prepares neo:latest and neo-light:latest.
6. Open NeoOptimize from the Start Menu or desktop shortcut.
7. Start with AI Doctor, Safe Cleanup, or Benchmark.

Unsigned public installers can trigger SmartScreen or reputation warnings. NeoOptimize is preparing a SignPath-based signing workflow for public freeware releases. SmartScreen reputation can still take time to build even after signing.

If an older lab build made Windows Security too strict, open NeoOptimize and run the Defender Lab Recovery action, or run:

powershell -NoProfile -ExecutionPolicy Bypass -File .\NeoOptimize.ps1 -Action DefenderAuditMode

This keeps Microsoft Defender enabled and moves aggressive lab ASR, Controlled Folder Access, and Network Protection policies to AuditMode.

Package manager manifests are prepared under distribution/.

Until each registry accepts the package, use the GitHub Release download and verify the SHA-256 checksum.

Free code signing is planned through SignPath.io with certificates from the SignPath Foundation. The signing workflow is intentionally gated until the public repository contains the complete application source and installer build recipe required for reproducible signing.

See SIGNPATH.md for the signing plan.

NeoOptimize is released as a free public utility. Support helps Zenthralix-Lab fund testing hardware, build infrastructure, documentation, security review, release maintenance, and future free software projects for the community.

• Email: neooptimizeofficial@gmail.com
• Buy Me a Coffee: https://buymeacoffee.com/nol.eight
• Saweria: https://saweria.co/dtechtive
• Dana: https://ik.imagekit.io/dtechtive/Dana

Made with love at Zenthralix-Lab with Codex.

NeoOptimize is released under the Apache License 2.0.