Regression: Codex Desktop has no persistent context/token usage indicator
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 42/100
Research direction
The payload names no Desktop files or tests. Start by comparing the Desktop task/status controls with the existing CLI/TUI status UI, then trace where model, context, compaction, and token totals are exposed. Done means a persistent or opt-in Desktop indicator distinguishes context occupancy from cumulative totals and warns about compaction.
Written by the indexing model from the issue text.
Description
Summary
Codex Desktop no longer shows a persistent, user-visible context and token usage indicator comparable to the Codex CLI/TUI status display.
The information is operationally important in long-running tasks: users need to know the active model, current context-window pressure/remaining budget, cumulative input and output token totals, and when compaction is approaching or has occurred.
Environment
- Codex Desktop: 26.707.9981.0
- Codex CLI installed locally: 0.144.4
- OS: Windows
- Model available in the CLI: gpt-5.6-sol
Steps to reproduce
- Open Codex Desktop.
- Start or open any task.
- Look near the composer, model selector, and task/status controls.
- There is no persistent context/token usage indicator or tooltip.
Actual behavior
The Desktop UI does not expose the live context-window usage, remaining context, input/output token totals, or a persistent model/status display. The CLI/TUI provides this information through its status UI, but the Desktop app does not provide an equivalent passive display.
Expected behavior
Restore an opt-in or always-visible Desktop status indicator, ideally near the composer or model selector, with:
- Active model/provider
- Current context usage and remaining effective context window
- Cumulative input, cached-input, and output token totals
- Clear separation between current context-window occupancy and cumulative session/thread totals
- Compaction status or a warning before automatic compaction
This is not merely a cosmetic preference; without the indicator, users cannot reliably decide when to compact, fork, or start a new task before context pressure affects work quality.
Related reports
Similar reports were filed as #23794, #23591, and #18201. This report confirms that the problem persists in a newer Desktop build.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.5k
- Avg merge
- 1m
- Merged PRs (30d)
- 1k
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from openai/codex
-
enhancement remote
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
-
bug CLI windows-os
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
-
macOS sandbox blocks hw.optional.arm64 sysctl, causing Flutter to misdetect Apple Silicon as x64 Openbug CLI sandbox
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
-
bug CLI TUI
Difficulty 2/5 1-3 hours Newbie friendliness 90/100
-
CLI config enhancement skills
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 86/100
kwakseongjae/auto-hwp#319 ·
-
area:cli bug filter-quality good first issue priority:medium
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
-
Difficulty 1/5 Under an hour Newbie friendliness 72/100
bevyengine/bevy#25861 ·
-
comp-datalake
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
ClickHouse/ClickHouse#121222 ·
-
A-linter
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
oxc-project/oxc#26863 ·