Codex CLI 0.96.0: Thread Compaction & Rate Limits
Codex CLI 0.96.0 brings async thread compaction, websocket rate limit events, and unified execution on non-Windows platforms. Plus fixes for TUI notes, thread listing, and tool injection atomicity.
openai
News zu OpenAI, ChatGPT und GPT-Modellen
Codex CLI 0.95.0: Skills, Parallel Shells & macOS Launcher
Codex CLI 0.95.0 adds personal skill loading, parallel shell execution, macOS app launcher, and hardens Git safety. Thread IDs now injected into shell environment.
Codex CLI 0.94.0: Plan Mode & Skills Loading
Plan mode is now on by default, personality config is stable with auto-migration, and skills load from .agents/skills with nested folder support. Runtime metrics added for diagnostics.
Codex App for macOS: Desktop Agent Interface
Codex app for macOS brings desktop-native agent threading, Git integration, voice dictation, and double rate limits for paid plans. ChatGPT Free and Go get it included.
Codex Web Search Now Enabled by Default
Web search is now on by default in Codex CLI and IDE Extension. Choose between cached results, live data, or disable it entirely via config.
Codex Team Config: Shared Configuration v2026-01-23
Team Config groups shared Codex settings across repos and machines. Share defaults, rules, and skills with automatic precedence handling and admin enforcement.
Cline Now Supports OpenAI OAuth Sign-In
Cline now supports OAuth sign-in with OpenAI. Authenticate once, access all your ChatGPT subscription models, no API keys required. Available now.