PolicyLockLLM v1.2.0

Release Notes - v1.2.0

---

Added

Centralized AI Service Registry

Expanded AI Platform Coverage

Theme Support

Enhanced Localization

---

Changed

Dynamic Rule Generation

Registry Caching Layer

Browser Lifecycle Handling

Automatic Tab Refresh

Documentation Overhaul

---

Improved

Block Page Experience

Extension UI

Styling System

---

Technical Improvements

Background Service Refactoring

DeclarativeNetRequest Enhancements

Storage Management

---

Compatibility

• Chrome Extension Manifest V3
• Existing blocking configuration remains supported
• Existing category-based filtering remains supported

---

Full Changelog: v1.1...v1.2.0

v1.1

PolicyLockLLM v1.1

✅ Added

• LLM-type filtering: Users can now choose to block all, text, image, or video LLMs. The dropdown in main.html drives dynamic rules in background.js, ensuring only the selected domains are intercepted ([feat(llmTypes) · mrf0rtuna4/PolicyLockLLM@f8af81a · GitHub](f8af81a)).
• Internationalization (i18n): All UI strings use __MSG_…__ placeholders, loaded at runtime via a new i18n.js script. The manifest’s default_locale is set to "ru", paving the way for multi-language support ([feat(i18n) · mrf0rtuna4/PolicyLockLLM@9c5b523 · GitHub](9c5b523)).

🛠 Changed

• UI/UX refresh:
  • Adopted CSS variables for colors, radii, and transitions in styles/main.css and blocked.html.
  • Switched to the Inter font (via Google Fonts) for a modern, legible interface.
  • Improved layout: responsive widths (90 % containers), centered flex layouts, subtle shadows, and consistent spacing ([chore(styles) · mrf0rtuna4/PolicyLockLLM@711c5f0 · GitHub](711c5f0)).
• Manifest update: Bumped "version" to "1.1" and added "default_locale": "ru" ([feat(i18n) · mrf0rtuna4/PolicyLockLLM@9c5b523 · GitHub](9c5b523)).

🐞 Fixed

• Locker logic overhaul:
  • Replaced static rule installation with updateBlockingRules to add/remove declarative net‐request rules based on stored preferences.
  • Introduced secure password storage: SHA-256 hashing, verifyPassword() checks, and support for random “self-lock” password generation.
  • Persisted settings (blocking, hardBlock, passwordHash, llmType) in chrome.storage.local and reactively updated rules on change events ([fix(locker) · mrf0rtuna4/PolicyLockLLM@990ce92 · GitHub](990ce92)).

---

Thank you for using PolicyLockLLM! Feel free to report any issues or suggest further improvements.

Full Changelog: v1.0...v1.1

v1.0

PolicyLockLLM v1

New Opportunities

1. Blocking LLM services
   • According to the list of domains BlockedDomains' in background.js` stops access to Chatgpt, Bard, Claude, Gemini, Poe and other popular llm sites.
2. Caste lock page
   • Instead of a standard Chrome error, Blocked is shown. html with a proprietary design and styles from the `styleles folder.
3. Automatic generation of rules
   • The structure of the `scripts' folder is provided for a simplified future addition and update of domains.
4. Hard block mode
   • the lock is not disconnected without entering an administrative password through the Popup interface,
     which provides increased safety for corporate use.

Technical characteristics

• manifest v3: fully compatible with current versions of Chrome, uses service volorkers and new API extensions.
• The composition of the assembly: BackGround.js, manifest.json, blocked.html, main.html,
  As well as the scripts/ styleles folders for modularity and expansion.

known restrictions

• The list of blocked domains requires manual update through editing Background.js.
• There is no centralized backend for managing admin-paralle and delimitation of access rights.

plans for the future

• Add UI to control the blocking rules without editing source code.
• implement integration with LDAP/Active Directory to authenticate administrators.
• automate the synchronization of the list of domains from the cloud repository.

Full Changelog: https://github.com/mrf0rtuna4/PolicyLockLLM/commits/v1.0