openai / openai/codex-plugin-cc
Stop-time review gate ALLOW results are silent/invisible in Claude Desktop's Code tab
Nobody has claimed this yet.
- Dominant language
- JavaScript
- Stars
- 33.3k
- Forks
- 2.3k
- PR merge metrics
- No merged PRs in 30d
Description
Environment
- Windows 11, Claude Desktop app, Code tab (not terminal CLI)
- Plugin
codex@openai-codexv1.0.6 (Project/shared scope) - Codex CLI
0.130.0-alpha.5, ChatGPT Plus login
Issue
With the stop-time review gate enabled, the gate runs correctly after nearly every turn — but ALLOW outcomes produce zero visible feedback in the chat (no message, no status change). This makes the feature look broken/inactive.
Running /codex:status shows it's actually working fine:
'''Latest finished:
task-msivzmtc-mfdhhw | completed | rescue | Codex Stop Gate Review
Summary: ALLOW: previous turn was only a status/explanation update, not edit-producing work.'''
Several completed ALLOW jobs were found this way, none visible in the transcript. In the terminal CLI, by contrast, a Ran 1 stop hook line with the result appears — so this seems specific to the Desktop Code tab.
Suggestion
Surface at least a minimal confirmation in chat when the Stop hook completes (e.g. Codex review: ALLOW), matching terminal CLI behavior — or document that Desktop results are only visible via /codex:status.
Contributor guide
No contributing guide indexed for this repository
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.
Research direction
Reproduce the stop-time review gate in Claude Desktop's Code tab using the stated plugin and Codex CLI versions, then compare it with the terminal CLI output. Start from the stop hook and the /codex:status entry point mentioned in the report. Done means an ALLOW result is visible in the Desktop chat, or the Desktop-only visibility is documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- javascript
- Domain
- desktop, tooling
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100