New Features
- Export complete TUI conversations to Markdown with
/export, either to the clipboard or a new file. (openai#37358) - Fork sessions with
codex exec fork, and archive or restore sessions from the TUI resume picker. (openai#37367, openai#37369, openai#37371) - Draft prompts while the TUI initializes, with resume and fork progress shown during startup. (openai#38642, openai#38788)
- View estimated thread credits or cost in
/status, status lines, and terminal titles for eligible workspaces. (openai#38281, openai#38282) - Use Amazon Bedrock Runtime as a built-in provider with AWS profile, region, and GPT-5.6 routing support. (openai#38470)
- Hooks can now run commands asynchronously and invoke MCP tools. (openai#37533, openai#38705)
Bug Fixes
- Model switches and settings updates no longer leave stale instructions behind or change an active turn midstream. (openai#37260, openai#38785)
- Resumed sessions now restore their persisted working directory and approval policy, with more accurate transcript previews. (openai#37198, openai#37368, openai#38605)
- Turns reconnect through temporary provider outages, and MCP servers recover after OAuth reauthentication without restarting Codex. (openai#37337, openai#37485, openai#38418)
- TUI startup no longer lets buffered terminal input activate prompts accidentally, and onboarding appears when authentication is missing. (openai#38641, openai#38643, openai#38644)
- Composer and transcript rendering now handle CRLF pastes, wrapped whitespace, and long URLs correctly. (openai#37709, openai#38380, openai#38704)
- Sandbox restrictions now fail closed for denied or unreadable paths across Linux and Windows. (openai#37875, openai#38026, openai#38416, openai#38660)
Documentation
- The bundled skill-creator guide is more focused, and validation now rejects unfinished TODO placeholders. (openai#38384)
Changelog
Full Changelog: openai/codex@rust-v0.147.0...rust-v0.148.0