Locrow AI 0.4.0 — Major AI Backend Overhaul
Release Date: November 2025
Minecraft Forge Mod: Locrow AI
Previous Version: 0.3.0 → New Version: 0.4.0
Overview
Locrow AI 0.4.0 represents a major architectural update to the AI backend system. This release focuses on security, improving backend reliability, modularity, and verification. It introduces a fully versioned manifest system, better Python environment management, and deeper integration between AI extensions, models, and the Minecraft mod context.
Core System Overhaul
- Rebranded all systems from AI Packages to AI Backend for clarity and structure.
- Moved core code and extensions into mod resources to unify deployment and improve sandboxing.
- Introduced a manifest system for backend versioning and verification, ensuring all backend files are hashed and signed.
- Added backend hash and seal verification for higher security and tamper resistance.
- Added load manager and runtime log system to track backend state and diagnostics.
- Implemented a setup log system for backend installation and initialization transparency.
Installation & Verification Improvements
- Introduced on-demand Python library installation ("install-what-you-need" system).
- Added model download and installation support with extraction and renaming.
- Combined wheel download and install into a single streamlined step.
- Added site-package verification and embedded wheel manifests inside the mod’s JAR resources.
- Added multi-pip installs per import name and automatic GPU version tagging for Python dependencies.
- Split backend verification into multiple progress stages for clearer feedback.
- Added AI backend verification on mod startup, checking whether any install tasks remain incomplete.
- Added install or load logic that intelligently decides when to reinitialize the backend.
- Fixed an issue that crashed the game on startup if your system was incompatible.
Security & Integrity
- Added a seal helper system for signing backend extensions.
- Added security checks to prevent unregistered extension declarations.
- Introduced a dynamic port system for backend communication isolation.
- Implemented manifest verification and backend signature validation to ensure integrity.
- Fixed an issue where Windows would ask for permission to allow the Python backend server to access the network.
Extension & Model Enhancements
- Added AI extension freeze functionality to stabilize runtime models.
- Added model registration and loading APIs.
- Added thread-locking features for model access and integrated them across bundled AI extensions.
- Introduced load model functions for streamlined AI extension workflows.
UI & User Experience
- Added third-party extensions warning and failed verification warning screens.
- Improved progress bar accuracy and multi-stage feedback during verification and installs.
- The AI Backend Manager screen now receives control after failed installs instead of reverting to the previous menu.
- Updated UI labels and buttons to reflect the new AI Backend terminology.
- Temporarily disabled the "Install" button on the AI Backend Manager Screen.
Developer Utilities
- Added registration hooks to tie AI scripts directly to mod contexts.
- Moved system probe functions into a dedicated
SystemProbe module.
- Improved exception handling across the entire backend and installation process.
- Added backend warmup and freeze APIs for testing and controlled startup.
Known Issues
- Minecraft may crash if the AI Backend verification process fails too quickly upon startup.
- The AI Backend may fail to close if Minecraft encounters certain types of crashes or if it is forcefully closed.
- Installation testing will fail if the RVC model "villager/model" is not installed (you can get around this issue by restarting Minecraft after it errors on the test stage of the installation process).
Summary
Locrow AI 0.4.0 is a foundational release that transforms the AI backend into a robust, secure, and extensible system. With manifest-based verification, improved model management, dynamic installation, and refined logging, it paves the way for a more modular and verifiable AI ecosystem within Minecraft.