Haleclipse/codex rust-v0.126.0-cometix on GitHub

New Features

Added persistent goal workflows with /goal, goal status indicators, pause/unpause/clear controls, and runtime continuation for long-running objectives (openai#18073, openai#18074, openai#18075, openai#18076, openai#18077).
Added configurable TUI key bindings through [tui.keymap], with shortcut hints reflecting the resolved config (openai#18593).
Added a top-level codex update command that runs the detected package-manager update path directly (openai#19933).
Expanded /plugins with marketplace installation, cached remote plugin bundles, remote uninstall support, and plugin-bundled hooks (openai#18704, openai#19456, openai#19705, openai#19914).
Added external-agent migration for recent sessions plus MCP servers, hooks, commands, and subagents (openai#19895, openai#19949).
Improved MultiAgentV2 configuration with feature-scoped thread caps, root/subagent context hints, configurable wait minimums, and nested spawn behavior independent of legacy depth limits (openai#19792, openai#19805, openai#20052, openai#20180).

Fixed turn/interrupt hanging when the target turn had already completed (openai#18392).
Thread resume now preserves the original model provider, handles large remote resume responses, and moves local cwd filtering into thread/list for faster resume/fork menus (openai#19287, openai#19920, openai#19931).
TUI scrollback now reflows after terminal resize, slash-command columns stay stable, shell mode exits cleanly with Esc, and keyboard reporting is reset on exit (openai#18575, openai#19511, openai#19625, openai#19986).
Bedrock GPT-5.4 now uses compatible apply-patch tooling, valid reasoning levels, and the current Mantle endpoint/model IDs (openai#19416, openai#19461, openai#20109).
Hardened managed network and sandbox boundaries around proxy bypasses, resolved connect targets, IPv6 scopes, Linux bridge helpers, Windows elevated pipes, and git -C approval checks (openai#19283, openai#19995, openai#19999, openai#20001, openai#20002, openai#20085).
Delegated patch approvals now show the proposed diff preview consistently with main-thread approvals (openai#19709).

Kept our release workflow (upstream added DMG/cosign/app-server bundles not needed for fork)
Kept cometix update_action.rs and updates.rs (npm-only update path)
Added TuiEvent::Resize handling to translate_overlay
Merged goal continuation logic with cometix rate limit notification in resume handler
GPT-5.5 now available in bundled models