Tags: pankaz/opencode-openai-codex-auth
Tags
Release v1.0.3 - Clean architecture and improved maintainability - **50% code reduction** in main file (512→256 lines) - **Modular design** with focused, single-responsibility modules - **Clean separation** of concerns across the codebase - `lib/logger.mjs` - Centralized debug logging - `lib/request-transformer.mjs` - Request body transformations - `lib/response-handler.mjs` - SSE to JSON conversion - Removed all debug console.logs - Improved readability and maintainability - Easier testing of individual components - Better organized imports and dependencies - Updated project structure in README - Added version pinning instructions - Improved module overview with new files - Removed bundled `codex-instructions.md` (fully migrated to GitHub fetching) All functionality remains identical - this is purely a refactoring release for better code organization and future maintainability. No breaking changes. ```bash npm install opencode-openai-codex-auth@1.0.3 ``` Or in your `opencode.json`: ```json { "plugin": ["opencode-openai-codex-auth"] } ``` Release v1.0.2 - Smart caching and stability improvements 🎯 Major Improvements: - ETag-based caching (304 Not Modified support) - Release tag tracking for API compatibility - Fixes "Instructions are not valid" errors 🐛 Bug Fixes: - Model normalization now defaults to gpt-5 - Fixed body.text initialization - Standardized error logging 📚 Documentation: - Added npm badges - Updated caching documentation
v1.0.0: Production-ready release Features: - ChatGPT Plus/Pro OAuth authentication with auto-refresh - Auto-updating Codex instructions from GitHub (24h cache) - Zero dependencies except @openauthjs/openauth - Modular architecture - Full tool support with automatic remapping - High reasoning effort with detailed thinking blocks