How I keep Codex side chats and old tasks findable without losing work
Losing a half-finished investigation because your terminal session could only hold one thread at a time is the real cost of scattered Codex context. The opt-in terminal workspace now supports multiple side-chat tabs plus searchable older tasks, so context survives restarts.
What it is
Codex adds multi-tab side chats in its opt-in terminal workspace, with saved local side chats reopened after restart. Older tasks are searchable from the agent command center by name, prompt preview, or working directory.
Applied case
A developer juggling a bug fix and an architecture question can open a side-chat tab for each thread, switch between them with the keyboard or mouse, and reopen them after restarting Codex because local side chats are saved. Later, they find an old task from the command center by searching its working directory or a prompt preview instead of scrolling history. Collapsing turn details in the full-screen transcript also folds completed activity under each prompt while keeping answers visible and tool output searchable.
- Switch tabs by keyboard or mouse; local side chats persist across restarts
When to use it
Use when you run parallel questions or need to revisit older tasks by name or directory.
Skip when you need to search arbitrary terminal history: task search covers tasks, not every past prompt outside them.
Read more
Recommended for you
- WorkflowCodex CLI
How I kept provider and pane control in Codex without losing flow
Replayed Codex conversations dropped reasoning summaries and steering messages could vanish mid-turn, making long sessions hard to trust.
- WorkflowCodex CLI
When to use incremental tool updates instead of legacy streaming?
Codex tool updates on Google Cloud Vertex AI previously arrived in batches rather than as they happen; incremental sending is now the default, with a flag to opt out.
- WorkflowVercel
When to use AI Gateway instead of direct image model APIs?
A separate MiniMax account, key, and bill just to generate product images slows you down until one gateway API key covers it.
- WorkflowVercel
How I filter agent tools without burning output tokens
Routing and policy checks with generative models spend tokens on text you never use.
Enjoyed this article?
Subscribe for new articles. No spam. Unsubscribe anytime.