trace-to-skill 0.1.94 → 0.1.96

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (26) hide show
  1. package/README.md +4 -4
  2. package/dist/src/benchmark.js +12 -0
  3. package/dist/src/benchmark.js.map +1 -1
  4. package/dist/src/demo.js +13 -0
  5. package/dist/src/demo.js.map +1 -1
  6. package/dist/src/init.js +3 -3
  7. package/dist/src/issueMap.js +12 -0
  8. package/dist/src/issueMap.js.map +1 -1
  9. package/dist/src/rules.js +35 -1
  10. package/dist/src/rules.js.map +1 -1
  11. package/dist/src/types.d.ts +1 -1
  12. package/docs/BENCHMARK.md +2 -0
  13. package/docs/CODEX_GITHUB_ISSUE_PAIN_MAP.md +25 -19
  14. package/docs/CODEX_ISSUE_MAP.md +3 -0
  15. package/docs/CODEX_ISSUE_RADAR.md +18 -19
  16. package/docs/DEMO.md +26 -21
  17. package/docs/FAILURE_TAXONOMY.md +16 -0
  18. package/docs/OPENAI_OSS_BRIEF.md +5 -5
  19. package/docs/SCORECARD.md +3 -1
  20. package/docs/USE_CASES.md +27 -1
  21. package/fixtures/codex-context-visibility.md +18 -0
  22. package/fixtures/codex-model-routing-mismatch.md +18 -0
  23. package/fixtures/github-codex-issues-export.json +82 -0
  24. package/llms.txt +4 -2
  25. package/package.json +10 -1
  26. package/schemas/analysis-result.schema.json +2 -0
package/docs/CODEX_ISSUE_MAP.md CHANGED
@@ -25,6 +25,7 @@ npx trace-to-skill lsp-audit . --format json
25
25
  | --- | --- | --- | --- |
26
26
  | Token burn, prompt-cache collapse, and usage drain | `tokens burning very fast`, `1% in 4 minutes`, `22 credits`, large cached input totals, `input_tokens` / `cached_input_tokens` / `cached_tokens` / `prompt_cache_key` rows, websocket reconnect cache drops, `write_stdin` empty polls, idle app usage, compaction tax, retry loops, subagent fan-out | `codex_token_burn` plus usage receipt and `prompt_cache_collapse` | `trace-to-skill usage-evidence ./usage-notes.md` first, then `trace-to-skill codex-report ./runs` for trace reports |
27
27
  | Usage bucket scope and percentage confusion | usage popover shows `Usage remaining`, `5h 97%`, `Weekly 95%`, reset date without explaining percent remaining vs percent used, rolling 7-day vs calendar-week, account-wide vs workspace/device/app/CLI scope | `codex_usage_bucket_confusion` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo usage-bucket-confusion` |
28
+ | Context/token indicator visibility | Desktop no longer shows visible context/token usage indicator, context-window pressure display or tooltip is missing, `/status` is not a passive replacement, users cannot tell when to compact or split long sessions | `codex_context_visibility` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo context-visibility` |
28
29
  | Usage reset schedule drift | weekly reset time changes, `reset_at` jumps, saved quota is wiped or pushed into the next window, outage compensation reset changes the anchor | `codex_usage_reset_drift` | `trace-to-skill usage-evidence ./usage-notes.md` or `trace-to-skill codex-report ./runs` |
29
30
  | Remote compact task failures | `/compact` or auto-compact fails, `responses/compact` stream disconnects, `timeout waiting for child process to exit`, `tcp_user_timeout` or `stream_idle_timeout_ms` workarounds, provider-id timeout drift | `codex_remote_compact` | `trace-to-skill codex-report ./runs` |
30
31
  | Windows helper and bundled tool path failures | bundled `rg.exe`, `node_repl.exe`, `codex-command-runner.exe`, Browser, Chrome, or Computer Use helpers resolve through `WindowsApps`, missing `%LOCALAPPDATA%\OpenAI\Codex\bin`, broken LocalCache helper bins, `CodexSandboxUsers` ACL gaps, EFS/copyfile failures | `codex_windows_helper_path` | `trace-to-skill codex-report ./runs` |
@@ -32,6 +33,7 @@ npx trace-to-skill lsp-audit . --format json
32
33
  | Process polling and high-CPU evidence | Windows `powershell.exe` / `pwsh` child process loops, `Get-CimInstance Win32_Process`, `Win32_PerfFormattedData_PerfProc_Process`, stale `chat_processes.json`, helper/renderer CPU samples | process audit receipt | `trace-to-skill process-audit ./process-notes.md` |
33
34
  | Tool-call integrity and rollback failures | `apply_patch` overwrites an existing `Add File` target, unmatched `tool_call_id`, `close_agent` hangs, failed revert/undo, unsafe diff application | `codex_tool_call_integrity` | `trace-to-skill codex-report ./runs` or `trace-to-skill guard-patch ./change.patch --root .` |
34
35
  | Undo, rewind, and pre-agent checkpoint needs | users want `/undo` or `/rewind`, double-Esc only rewinds chat state, untracked/gitignored files are not protected by commits, and manual recovery needs reviewable pre-agent evidence | workspace checkpoint bundle | `trace-to-skill checkpoint . --output .trace-to-skill/checkpoints/before-codex` before agent work |
36
+ | Model routing mismatch | selected `gpt-5.3-codex` in `config.toml`, TUI, or `--model`, but SSE `response.created` / `response.model` shows `gpt-5.2`, silent fallback, no warning, no fallback notice | `codex_model_routing_mismatch` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo model-routing-mismatch` |
35
37
  | Latency regressions | GPT-5.5 Fast feels like Standard, simple tasks take 10-20+ minutes, pre-first-token or thinking stalls, slow search/read/compaction, hours for small code changes | `codex_latency_regression` | `trace-to-skill codex-report ./runs` |
36
38
  | Thinking or stream hangs | accepted turn, completed local tool output, no streamed assistant follow-up, long gap before first `response_item`, `time.busy` milliseconds with `time.idle` hundreds of seconds, Stop/Ctrl+C cannot interrupt, subagent parent stuck | `codex_thinking_hang` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo thinking-hang` |
37
39
  | Clipboard, paste, and generated attachment regressions | `Copy as Markdown` missing, Copy menu only exports metadata, long pasted prompts become `Pasted text.txt`, generated attachments cannot preview/edit/revert, `/goal` ignores non-empty fileAttachments | `codex_clipboard_attachment` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo clipboard-attachment` |
@@ -54,6 +56,7 @@ npx trace-to-skill lsp-audit . --format json
54
56
  | LSP auto-detect readiness | Codex users want language-aware navigation, diagnostics, references, rename, or install guidance before edits | language-server metadata | `trace-to-skill lsp-audit . --format json` |
55
57
  | Context compaction failures | `Error running remote compact task`, `context_length_exceeded`, compaction loops, `responses/compact` stream disconnects | `context_compaction` | `trace-to-skill analyze ./runs` |
56
58
  | Latest-turn drift | Codex answers an older prompt, repeats a previous response, redoes an already fixed task, forgets recent edits after compaction, or leaks raw tool payload text | `codex_latest_turn_drift` | `trace-to-skill codex-report ./runs` |
59
+ | Model routing mismatch | selected model differs from actual server-side `response.model`, silent fallback, or missing model-unavailable warning | `codex_model_routing_mismatch` | `trace-to-skill codex-report ./runs` or `trace-to-skill demo model-routing-mismatch` |
57
60
  | Session resume, project history, and state failures | `codex resume` picker freezes, Desktop project/search/sidebar hides existing threads, large rollout JSONL, short or stale `session_index.jsonl`, transcript-like sidebar title chunks, unindexed rollout thread ids, `Could not load archived chats`, `state_5.sqlite`, `thread_goals` | `codex_session_state` | `trace-to-skill codex-report ./runs` or `trace-to-skill session-audit ~/.codex --format json` |
58
61
  | Sandbox and permission blockers | Windows sandbox setup refresh, `os error 740`, ACL/ownership drift, approval-mode mismatch | `sandbox_permission` | `trace-to-skill analyze ./runs` |
59
62
  | Auth and connectivity failures | `token_exchange_failed`, `auth.openai.com/oauth/token`, missing CA certificates, proxy/TLS, IPv6, Cloudflare, stream disconnects | `codex_connectivity` | `trace-to-skill codex-report ./runs` |
package/docs/CODEX_ISSUE_RADAR.md CHANGED
@@ -1,10 +1,10 @@
1
1
  # GitHub Issue Pain Map
2
2
 
3
- Generated: 2026-06-01T02:48:00.507Z
3
+ Generated: 2026-06-01T03:13:20.215Z
4
4
 
5
5
  Issues analyzed: **46**
6
- Matched issues: **18**
7
- Unmatched issues: **28**
6
+ Matched issues: **20**
7
+ Unmatched issues: **26**
8
8
 
9
9
  This report maps GitHub issues onto deterministic `trace-to-skill` failure classes. Fetch a repository directly with `--repo`, or export issues with `gh issue list` / `gh search issues` and pass the JSON file.
10
10
 
@@ -20,17 +20,17 @@ gh issue list --repo openai/codex --state all --limit 100 --json number,title,bo
20
20
  | Priority | Kind | Severity | Issues | Comments | Reactions | Example |
21
21
  | ---: | --- | --- | ---: | ---: | ---: | --- |
22
22
  | 2438 | `codex_token_burn` | high | 4 | 1151 | 620 | [#14593 Burning tokens very fast](https://github.com/openai/codex/issues/14593) |
23
- | 2221 | `weak_evidence` | medium | 46 | 4755 | 7792 | [#14593 Burning tokens very fast](https://github.com/openai/codex/issues/14593) |
23
+ | 2221 | `weak_evidence` | medium | 46 | 4755 | 7794 | [#14593 Burning tokens very fast](https://github.com/openai/codex/issues/14593) |
24
24
  | 884 | `sensitive_file_access` | high | 1 | 75 | 396 | [#2847 A way to exclude sensitive files](https://github.com/openai/codex/issues/2847) |
25
25
  | 805 | `codex_auth_verification` | high | 3 | 436 | 166 | [#20161 Phone number verification doesn't work](https://github.com/openai/codex/issues/20161) |
26
+ | 631 | `codex_context_visibility` | high | 1 | 160 | 227 | [#23794 Codex Desktop no longer shows visible context/token usage indicator](https://github.com/openai/codex/issues/23794) |
26
27
  | 442 | `codex_tool_call_integrity` | high | 1 | 61 | 182 | [#2998 IDE-integrated diff / approval](https://github.com/openai/codex/issues/2998) |
27
28
  | 376 | `codex_remote_compact` | high | 2 | 147 | 101 | [#14860 Error running remote compact task](https://github.com/openai/codex/issues/14860) |
28
29
  | 376 | `context_compaction` | high | 2 | 147 | 101 | [#14860 Error running remote compact task](https://github.com/openai/codex/issues/14860) |
29
30
  | 351 | `codex_terminal_output_integrity` | high | 1 | 66 | 134 | [#2558 Codex client output truncated when scrolling in Zellij](https://github.com/openai/codex/issues/2558) |
31
+ | 324 | `codex_model_routing_mismatch` | high | 1 | 169 | 69 | [#11189 GPT-5.3-Codex being routed to GPT-5.2](https://github.com/openai/codex/issues/11189) |
30
32
  | 261 | `premature_completion` | high | 1 | 60 | 92 | [#2448 Codex CLI: Plus users hitting usage limits extremely quickly compared to competitors](https://github.com/openai/codex/issues/2448) |
31
33
  | 247 | `codex_connectivity` | high | 2 | 192 | 14 | [#12764 The codex cli giving: 401 unauthorized](https://github.com/openai/codex/issues/12764) |
32
- | 181 | `codex_latest_turn_drift` | high | 1 | 58 | 53 | [#8648 Codex replies to earlier messages instead of latest one in conversations](https://github.com/openai/codex/issues/8648) |
33
- | 168 | `codex_resource_leak` | high | 1 | 97 | 27 | [#10432 High GPU usage (70–90%) on macOS with Codex app](https://github.com/openai/codex/issues/10432) |
34
34
 
35
35
  ## Maintainer Roadmap
36
36
 
@@ -39,8 +39,8 @@ gh issue list --repo openai/codex --state all --limit 100 --json number,title,bo
39
39
  | 1 | Usage evidence fixture and support-ready token report | 4 issue(s), 1151 comment(s), severity high; top signal: codex_token_burn. | `trace-to-skill usage-evidence ./usage-notes.md --output usage-evidence.md` |
40
40
  | 2 | Privacy/safety guardrail and redacted support bundle | 1 issue(s), 75 comment(s), severity high; top signal: sensitive_file_access. | `trace-to-skill diagnostics-bundle ~/.codex --output codex-diagnostics` |
41
41
  | 3 | Auth verification fixture and login support report | 3 issue(s), 436 comment(s), severity high; top signal: codex_auth_verification. | `trace-to-skill codex-report ./runs --output openai-codex-auth-issue.md` |
42
- | 4 | Patch safety fixture and pre-agent checkpoint workflow | 1 issue(s), 61 comment(s), severity high; top signal: codex_tool_call_integrity. | `trace-to-skill checkpoint . --output .trace-to-skill/checkpoints/before-codex` |
43
- | 5 | Compaction/session regression fixture and Codex issue report | 2 issue(s), 147 comment(s), severity high; top signal: codex_remote_compact. | `trace-to-skill codex-report ./runs --output openai-codex-issue.md` |
42
+ | 4 | Context visibility fixture and Desktop UI evidence report | 1 issue(s), 160 comment(s), severity high; top signal: codex_context_visibility. | `trace-to-skill codex-report ./runs --output openai-codex-context-visibility.md` |
43
+ | 5 | Patch safety fixture and pre-agent checkpoint workflow | 1 issue(s), 61 comment(s), severity high; top signal: codex_tool_call_integrity. | `trace-to-skill checkpoint . --output .trace-to-skill/checkpoints/before-codex` |
44
44
 
45
45
  ## Suggested Next Actions
46
46
 
@@ -78,33 +78,30 @@ Example issues:
78
78
  Evidence rule prompts:
79
79
  - When reporting Codex sign-in or account-verification failures, capture the Codex app/CLI/extension version, surface, OS, account type without secrets, workspace or organization context, SSO provider, whether the flow is ChatGPT sign-in, phone/SMS/OTP verification, or extension chat initialization, exact redacted error text, timestamps, whether another device/browser/account works, logout/login attempts, and screenshots with phone numbers, tokens, and email addresses redacted.
80
80
 
81
- ### codex_tool_call_integrity
81
+ ### codex_context_visibility
82
82
 
83
- Priority score: 442. 1 issue(s), 61 comment(s).
83
+ Priority score: 631. 1 issue(s), 160 comment(s).
84
84
 
85
85
  Example issues:
86
- - [#2998 IDE-integrated diff / approval](https://github.com/openai/codex/issues/2998) (61 comments; labels: enhancement, extension)
86
+ - [#23794 Codex Desktop no longer shows visible context/token usage indicator](https://github.com/openai/codex/issues/23794) (160 comments; labels: bug, context, app)
87
87
 
88
88
  Evidence rule prompts:
89
- - When reporting Codex tool-call integrity failures, capture the exact tool input and output, app/CLI/extension version, OS/IDE, workspace git state, affected file path and whether it already existed or was a symlink, diff before/after, tool_call_id sequence, durable thread state for subagents, rollback/revert attempts, and whether a clean repo reproduction fails the same way.
89
+ - When reporting Codex context-visibility regressions, capture Codex Desktop version, OS, surface, screenshot or short recording of the chat input area, whether the prior context/token indicator or tooltip was visible before the update, exact UI route where it disappeared, local session metadata showing context/window pressure if available, `/status` output if relevant, compaction timing, whether CLI/TUI still exposes a statusline, and how the missing indicator affects long-session decisions.
90
90
 
91
- ### codex_remote_compact
91
+ ### codex_tool_call_integrity
92
92
 
93
- Priority score: 376. 2 issue(s), 147 comment(s).
93
+ Priority score: 442. 1 issue(s), 61 comment(s).
94
94
 
95
95
  Example issues:
96
- - [#14860 Error running remote compact task](https://github.com/openai/codex/issues/14860) (90 comments; labels: bug, context)
97
- - [#9544 Error running remote compact task: stream disconnected before completion](https://github.com/openai/codex/issues/9544) (57 comments; labels: bug, context)
96
+ - [#2998 IDE-integrated diff / approval](https://github.com/openai/codex/issues/2998) (61 comments; labels: enhancement, extension)
98
97
 
99
98
  Evidence rule prompts:
100
- - When reporting Codex remote compact failures, capture app/CLI/extension version, OS, model and reasoning/speed mode, provider config without secrets, exact /compact or auto-compact error, `responses/compact` endpoint shape, timeout values such as tcp_user_timeout or stream_idle_timeout_ms, context/token level before compaction, whether lowering reasoning/speed changes behavior, whether local fallback or a new session recovers, and related thread/feedback ids.
99
+ - When reporting Codex tool-call integrity failures, capture the exact tool input and output, app/CLI/extension version, OS/IDE, workspace git state, affected file path and whether it already existed or was a symlink, diff before/after, tool_call_id sequence, durable thread state for subagents, rollback/revert attempts, and whether a clean repo reproduction fails the same way.
101
100
 
102
101
  ## Unmatched Issues
103
102
 
104
103
  - [#10410 Codex Desktop App: macOS Intel (x86_64) support](https://github.com/openai/codex/issues/10410) (190 comments; labels: enhancement, app)
105
104
  - [#10450 Remote Development in Codex Desktop App](https://github.com/openai/codex/issues/10450) (176 comments; labels: enhancement, app)
106
- - [#11189 GPT-5.3-Codex being routed to GPT-5.2](https://github.com/openai/codex/issues/11189) (169 comments; labels: bug, CLI)
107
- - [#23794 Codex Desktop no longer shows visible context/token usage indicator](https://github.com/openai/codex/issues/23794) (160 comments; labels: bug, context, app)
108
105
  - [#14048 All models — Codex CLI hangs indefinitely on all prompts, no response generated](https://github.com/openai/codex/issues/14048) (131 comments; labels: bug, agent)
109
106
  - [#2604 Subagent Support](https://github.com/openai/codex/issues/2604) (103 comments; labels: enhancement, subagent)
110
107
  - [#12564 Allow renaming task/thread titles to improve history navigation](https://github.com/openai/codex/issues/12564) (77 comments; labels: enhancement, extension)
@@ -113,3 +110,5 @@ Evidence rule prompts:
113
110
  - [#2796 BUG: VSCode IDE Plugin on SSH Connection: "Failed to load tasks."](https://github.com/openai/codex/issues/2796) (71 comments; labels: bug, extension)
114
111
  - [#16231 High CPU usage on macOS after updating Codex in VS Code extension to 26.325.31654](https://github.com/openai/codex/issues/16231) (71 comments; labels: bug, extension, regression, performance)
115
112
  - [#7156 Codex hangs during cli command execution](https://github.com/openai/codex/issues/7156) (70 comments; labels: bug, CLI)
113
+ - [#4313 Extension for JetBrains IDEs (PyCharm, IntelliJ, etc.)](https://github.com/openai/codex/issues/4313) (70 comments; labels: enhancement)
114
+ - [#13041 WebSocket upgrade succeeds then server closes with 1008 Policy (falls back to HTTPS)](https://github.com/openai/codex/issues/13041) (70 comments; labels: bug, connectivity)
package/docs/DEMO.md CHANGED
@@ -1,10 +1,10 @@
1
1
  # trace-to-skill Demo
2
2
 
3
- Scenario: **Codex sign-in and account verification failure**
3
+ Scenario: **Codex context or token usage indicator missing**
4
4
 
5
- Phone verification, ChatGPT sign-in account routing, or extension chat initialization blocks Codex before a usable session starts.
5
+ Desktop context or token usage indicators disappear, leaving long-session compaction pressure invisible.
6
6
 
7
- Fixture: `fixtures/codex-auth-verification.md`
7
+ Fixture: `fixtures/codex-context-visibility.md`
8
8
 
9
9
  This is a packaged public fixture, so you can try the project without collecting a private trace first.
10
10
 
@@ -14,7 +14,7 @@ This is a packaged public fixture, so you can try the project without collecting
14
14
 
15
15
  Score: **75/100**
16
16
 
17
- Likely failure class: **Codex sign-in or account verification failure (codex_auth_verification, high)**
17
+ Likely failure class: **Codex context or token usage indicator missing (codex_context_visibility, high)**
18
18
 
19
19
  Agent workflow needs clearer verification, instruction, or security hardening before broad reuse.
20
20
 
@@ -23,24 +23,25 @@ Agent workflow needs clearer verification, instruction, or security hardening be
23
23
  ```md
24
24
  ### What happened?
25
25
 
26
- trace-to-skill detected Codex sign-in or account verification failure (codex_auth_verification). Codex first-party sign-in, phone verification, account-type routing, and extension chat initialization failures block users before they can produce useful debugging traces; reports need account surface and verification evidence without exposing tokens or phone numbers.
26
+ trace-to-skill detected Codex context or token usage indicator missing (codex_context_visibility). Desktop users need passive context-pressure visibility during long coding sessions so they can decide when to compact, split a thread, reduce pasted context, or avoid context loss before the app forces compaction.
27
27
 
28
28
  ### Detected failure class
29
29
 
30
- - codex_auth_verification: Codex sign-in or account verification failure (high)
30
+ - codex_context_visibility: Codex context or token usage indicator missing (high)
31
31
 
32
32
  ### Evidence
33
33
 
34
- #### Codex sign-in or account verification failure
35
- - fixtures/codex-auth-verification.md:5 - - Phone number verification doesn't work after logging out on one device and signing in on another device with SSO.
36
- - fixtures/codex-auth-verification.md:6 - - The Codex sign-in screen asks for phone verification even though the ChatGPT account normally uses Google or Apple SSO.
37
- - fixtures/codex-auth-verification.md:7 - - The SMS verification code is not received, shows `invalid_phone_number`, or the user gets a phone call from random numbers instead of a predictable verification code.
38
- - fixtures/codex-auth-verification.md:8 - - "Sign in With ChatGPT" needs to be robust across Plus, Pro, Teams, Enterprise, personal workspace, and organization-verified account types.
39
- - fixtures/codex-auth-verification.md:10 - - In the VS Code extension, a new chat shows "Error starting conversation" while initializing a chat after sign-in.
34
+ #### Codex context or token usage indicator missing
35
+ - fixtures/codex-context-visibility.md:5 - - Codex Desktop no longer shows a visible context/token usage indicator in the chat UI after an update.
36
+ - fixtures/codex-context-visibility.md:6 - - Previously the app exposed context usage information, context-window pressure, or a tooltip near the input area.
37
+ - fixtures/codex-context-visibility.md:8 - - `/status` is useful as an explicit command, but it is not a replacement for passive context awareness during long-running desktop threads.
38
+ - fixtures/codex-context-visibility.md:9 - - The missing indicator affects professional coding workflows because users cannot tell when to compact, start a new thread, reduce pasted context, or split work before context loss.
39
+ - fixtures/codex-context-visibility.md:10 - - A related report says the data exists in local session logs, but the app no longer exposes it passively.
40
+ - fixtures/codex-context-visibility.md:15 - - Screenshot or short recording of the chat input area where the context indicator or tooltip used to appear.
40
41
 
41
42
  ### Diagnostics to attach
42
43
 
43
- - When reporting Codex sign-in or account-verification failures, capture the Codex app/CLI/extension version, surface, OS, account type without secrets, workspace or organization context, SSO provider, whether the flow is ChatGPT sign-in, phone/SMS/OTP verification, or extension chat initialization, exact redacted error text, timestamps, whether another device/browser/account works, logout/login attempts, and screenshots with phone numbers, tokens, and email addresses redacted.
44
+ - When reporting Codex context-visibility regressions, capture Codex Desktop version, OS, surface, screenshot or short recording of the chat input area, whether the prior context/token indicator or tooltip was visible before the update, exact UI route where it disappeared, local session metadata showing context/window pressure if available, `/status` output if relevant, compaction timing, whether CLI/TUI still exposes a statusline, and how the missing indicator affects long-session decisions.
44
45
 
45
46
  ### Privacy
46
47
 
@@ -49,22 +50,23 @@ trace-to-skill detected Codex sign-in or account verification failure (codex_aut
49
50
 
50
51
  ## Findings
51
52
 
52
- ### 1. Codex sign-in or account verification failure
53
+ ### 1. Codex context or token usage indicator missing
53
54
 
54
55
  Severity: **high**
55
56
 
56
- Codex first-party sign-in, phone verification, account-type routing, and extension chat initialization failures block users before they can produce useful debugging traces; reports need account surface and verification evidence without exposing tokens or phone numbers.
57
+ Desktop users need passive context-pressure visibility during long coding sessions so they can decide when to compact, split a thread, reduce pasted context, or avoid context loss before the app forces compaction.
57
58
 
58
59
  Evidence:
59
- - `fixtures/codex-auth-verification.md:5` - Phone number verification doesn't work after logging out on one device and signing in on another device with SSO.
60
- - `fixtures/codex-auth-verification.md:6` - The Codex sign-in screen asks for phone verification even though the ChatGPT account normally uses Google or Apple SSO.
61
- - `fixtures/codex-auth-verification.md:7` - The SMS verification code is not received, shows `invalid_phone_number`, or the user gets a phone call from random numbers instead of a predictable verification code.
62
- - `fixtures/codex-auth-verification.md:8` - "Sign in With ChatGPT" needs to be robust across Plus, Pro, Teams, Enterprise, personal workspace, and organization-verified account types.
63
- - `fixtures/codex-auth-verification.md:10` - In the VS Code extension, a new chat shows "Error starting conversation" while initializing a chat after sign-in.
60
+ - `fixtures/codex-context-visibility.md:5` - Codex Desktop no longer shows a visible context/token usage indicator in the chat UI after an update.
61
+ - `fixtures/codex-context-visibility.md:6` - Previously the app exposed context usage information, context-window pressure, or a tooltip near the input area.
62
+ - `fixtures/codex-context-visibility.md:8` - `/status` is useful as an explicit command, but it is not a replacement for passive context awareness during long-running desktop threads.
63
+ - `fixtures/codex-context-visibility.md:9` - The missing indicator affects professional coding workflows because users cannot tell when to compact, start a new thread, reduce pasted context, or split work before context loss.
64
+ - `fixtures/codex-context-visibility.md:10` - A related report says the data exists in local session logs, but the app no longer exposes it passively.
65
+ - `fixtures/codex-context-visibility.md:15` - Screenshot or short recording of the chat input area where the context indicator or tooltip used to appear.
64
66
 
65
67
  Suggested rule:
66
68
 
67
- > When reporting Codex sign-in or account-verification failures, capture the Codex app/CLI/extension version, surface, OS, account type without secrets, workspace or organization context, SSO provider, whether the flow is ChatGPT sign-in, phone/SMS/OTP verification, or extension chat initialization, exact redacted error text, timestamps, whether another device/browser/account works, logout/login attempts, and screenshots with phone numbers, tokens, and email addresses redacted.
69
+ > When reporting Codex context-visibility regressions, capture Codex Desktop version, OS, surface, screenshot or short recording of the chat input area, whether the prior context/token indicator or tooltip was visible before the update, exact UI route where it disappeared, local session metadata showing context/window pressure if available, `/status` output if relevant, compaction timing, whether CLI/TUI still exposes a statusline, and how the missing indicator affects long-session decisions.
68
70
 
69
71
 
70
72
  ## Reporter Notes
@@ -83,10 +85,12 @@ Suggested rule:
83
85
  - `windows-helper-path`: Windows Desktop exposes bundled rg/node/plugin helpers from WindowsApps or missing LocalCache paths that cannot execute.
84
86
  - `approval-friction`: Repeated approval prompts, Approve for this session misses, and noisy trusted MCP tool approvals.
85
87
  - `latency-regression`: Fast mode feels like Standard, with long thinking, search, read, or compaction stalls.
88
+ - `model-routing-mismatch`: Codex shows one selected model while SSE response evidence shows a different server-side model was used.
86
89
  - `thinking-hang`: A turn or tool call completes, but the session stays on Thinking or Working with no streamed follow-up.
87
90
  - `clipboard-attachment`: Copy as Markdown, long-paste conversion, or generated Pasted text.txt attachments break prompt and report workflows.
88
91
  - `deeplink-launch`: OAuth callbacks, notification clicks, mobile links, or `codex app <path>` external activation fail to route into Codex.
89
92
  - `connector-auth-cache`: App connectors keep stale `link_*` auth or discovery metadata after reauth-required responses.
93
+ - `auth-verification`: Phone verification, ChatGPT sign-in account routing, or extension chat initialization blocks Codex before a usable session starts.
90
94
  - `mcp-discovery-mismatch`: MCP servers work in CLI or one config scope but are absent in Desktop, VS Code, WSL, or project-local sessions.
91
95
  - `mcp-streamable-http`: Streamable HTTP or SSE MCP servers pass initialize or tools/list but fail parsing, handshakes, auth gating, stale sessions, or reconnects.
92
96
  - `hooks-runtime`: Hooks duplicate, stop firing, warn about stale config, skip surfaces, or become hard to manage in Desktop settings.
@@ -117,6 +121,7 @@ trace-to-skill demo hooks-runtime
117
121
  trace-to-skill demo terminal-output-integrity
118
122
  trace-to-skill demo subagent-lifecycle
119
123
  trace-to-skill demo usage-bucket-confusion
124
+ trace-to-skill demo context-visibility
120
125
  trace-to-skill demo file-tree-ui
121
126
  trace-to-skill demo usage-reset-drift
122
127
  ```
package/docs/FAILURE_TAXONOMY.md CHANGED
@@ -32,6 +32,14 @@ Common signals include `responds to an earlier message`, `ignoring my latest mes
32
32
 
33
33
  The fix is to capture app/CLI/extension version, model and reasoning effort, context-window percent or token counts, compaction timing, the exact latest user request, the stale earlier request or response it answered instead, thread or feedback id, whether resending the same message fixes it, and any raw internal tool payload leaked into the chat UI.
34
34
 
35
+ ## Codex Selected Model Differs From Actual Routed Model
36
+
37
+ Codex shows one selected model in config, TUI, CLI flags, or UI, but the actual server-side response uses a different model. This is different from generic latency: the key evidence is a selected-model versus `response.model` mismatch or silent fallback.
38
+
39
+ Common signals include `GPT-5.3-Codex being routed to GPT-5.2`, `config.toml` and TUI set to `gpt-5.3-codex` while SSE `response.created` shows `response.model=gpt-5.2-2025-12-11`, `RUST_LOG=codex_api::sse::responses=trace`, `codex exec --model gpt-5.3-codex`, no warning, no fallback notice, and verification briefly restoring access before a silent reroute.
40
+
41
+ The fix is to capture Codex app/CLI/extension version, subscription/workspace, selected model from `config.toml`, TUI, command flag, or UI, actual server-side model from SSE `response.created` / `response.model`, the exact `RUST_LOG` or trace command used, timestamp, account or verification state without secrets, whether API and Codex routes differ, whether a warning/fallback notice appeared, and a minimal one-prompt reproduction with redacted logs.
42
+
35
43
  ## Codex Latency Regression
36
44
 
37
45
  Codex can regress from fast interactive work into long pre-first-token stalls, extended thinking, slow read/search orchestration, compaction delays, or model routing that makes a fast mode feel like a standard or higher-reasoning mode.
@@ -144,6 +152,14 @@ Common signals include `Usage remaining`, compact rows like `5h 97%` and `Weekly
144
152
 
145
153
  The fix is to capture subscription plan, account/workspace, app/CLI version, surface, timestamp, screenshot or redacted popover text, 5h percentage, weekly percentage, reset time/date, whether values are used or remaining, whether weekly is rolling or calendar-based, whether weekly includes app/CLI/cloud/review usage, `/status` output, usage dashboard state, and whether other devices or workspaces show the same values.
146
154
 
155
+ ## Codex Context Or Token Usage Indicator Missing
156
+
157
+ Codex Desktop can regress by hiding or removing passive context/token usage visibility from the chat UI. This is different from confusing quota percentages: the issue is that users cannot see context-window pressure at all during long desktop threads.
158
+
159
+ Common signals include `visible context/token usage indicator`, `context-window pressure`, missing context usage tooltip near the input area, `Context N% used` or `Context N% remaining` requests, `/status` not being a replacement for passive context awareness, local session logs still containing context data, and users being unable to decide when to compact, split a thread, reduce pasted context, or avoid context loss.
160
+
161
+ The fix is to capture Codex Desktop version, OS, surface, screenshot or short recording of the chat input area, whether the prior context/token indicator or tooltip was visible before the update, exact UI route where it disappeared, local session metadata showing context/window pressure if available, `/status` output if relevant, compaction timing, whether CLI/TUI still exposes a statusline, and how the missing indicator affects long-session decisions.
162
+
147
163
  ## Codex Subagent Prompt Leakage
148
164
 
149
165
  Codex MultiAgentV2 child agents can fail the task boundary even when the parent asks for isolated children. When `spawn_agent` with `fork_turns: "none"` records the delegated task as an assistant/commentary JSON envelope, or a same-turn parallel child sees a sibling prompt, independent review, QA, and security lanes are no longer independent.
package/docs/OPENAI_OSS_BRIEF.md CHANGED
@@ -3,14 +3,14 @@
3
3
  | Field | Value |
4
4
  | --- | --- |
5
5
  | Repository | https://github.com/grnbtqdbyx-create/trace-to-skill |
6
- | Package | trace-to-skill@0.1.94 |
6
+ | Package | trace-to-skill@0.1.96 |
7
7
  | License | Apache-2.0 |
8
8
  | Codex readiness | ready (100/100) |
9
- | Benchmark | pass, 39 cases |
9
+ | Benchmark | pass, 41 cases |
10
10
 
11
11
  ## Why This Repository Qualifies
12
12
 
13
- trace-to-skill helps open-source maintainers adopt Codex safely by turning failed coding-agent runs into evidence-backed rules, reusable workflows, CI gates, and a weekly Codex Issue Radar for live GitHub issue demand. It supports real maintenance work: PR review, issue triage, release quality, MCP risk, prompt-injection defense, privacy-preserving trace sharing, and repeat failure reduction. The repository is ready, scores 100/100 on the local Codex readiness doctor, and ships a deterministic benchmark with 39 public fixture cases.
13
+ trace-to-skill helps open-source maintainers adopt Codex safely by turning failed coding-agent runs into evidence-backed rules, reusable workflows, CI gates, and a weekly Codex Issue Radar for live GitHub issue demand. It supports real maintenance work: PR review, issue triage, release quality, MCP risk, prompt-injection defense, privacy-preserving trace sharing, and repeat failure reduction. The repository is ready, scores 100/100 on the local Codex readiness doctor, and ships a deterministic benchmark with 41 public fixture cases.
14
14
 
15
15
  ### 500-Character Version
16
16
 
@@ -27,10 +27,10 @@ API credits would power optional maintainer workflows on top of the local determ
27
27
  ## Evidence
28
28
 
29
29
  - Public repository: https://github.com/grnbtqdbyx-create/trace-to-skill
30
- - One-command package: npx trace-to-skill@0.1.94
30
+ - One-command package: npx trace-to-skill@0.1.96
31
31
  - Open-source license: Apache-2.0
32
32
  - Codex readiness doctor: ready, 100/100, 0 failed checks.
33
- - Public fixture benchmark: pass, 39 cases.
33
+ - Public fixture benchmark: pass, 41 cases.
34
34
  - GitHub issue demand mining: issue-map fetches or reads piped GitHub CLI issue JSON, then ranks OpenAI/Codex issues by failure class, comments, reactions, evidence gaps, and Maintainer Roadmap next artifacts.
35
35
  - Weekly Codex Issue Radar: init --issue-map-repo owner/name scaffolds a scheduled Action that fetches live GitHub issues and publishes the pain map to the job summary or a stable tracking issue comment.
36
36
  - Maintainer control: generated rules are suggestions, evidence is line-linked, and secrets can be redacted before sharing.
package/docs/SCORECARD.md CHANGED
@@ -9,7 +9,7 @@ Status: **pass**
9
9
  | Failed doctor checks | 0 |
10
10
  | Critical findings | 0 |
11
11
  | Built-in benchmark | pass |
12
- | Benchmark cases | 39 |
12
+ | Benchmark cases | 41 |
13
13
 
14
14
  ## Doctor Summary
15
15
 
@@ -31,6 +31,7 @@ This benchmark runs the public fixture pack that ships with the repository and p
31
31
  | Codex conversation fork context bloat | `fixtures/codex-context-fork-bloat.md` | 59 | 3 | 0 | `codex_context_fork_bloat`, `codex_thinking_hang`, `weak_evidence` | pass |
32
32
  | Codex subagent prompt leakage or boundary failure | `fixtures/codex-subagent-prompt-leakage.md` | 75 | 2 | 0 | `codex_subagent_prompt_leakage`, `weak_evidence` | pass |
33
33
  | Codex latest-turn drift after compaction | `fixtures/codex-latest-turn-drift.md` | 59 | 3 | 0 | `codex_latest_turn_drift`, `premature_completion`, `weak_evidence` | pass |
34
+ | Codex selected model differs from actual routed model | `fixtures/codex-model-routing-mismatch.md` | 75 | 2 | 0 | `codex_model_routing_mismatch`, `weak_evidence` | pass |
34
35
  | Codex model and runtime latency regression | `fixtures/codex-latency-regression.md` | 75 | 2 | 0 | `codex_latency_regression`, `weak_evidence` | pass |
35
36
  | Codex thinking and stream hang | `fixtures/codex-thinking-hang.md` | 75 | 2 | 0 | `codex_thinking_hang`, `weak_evidence` | pass |
36
37
  | Codex clipboard, paste, and attachment workflow regression | `fixtures/codex-clipboard-attachment.md` | 75 | 2 | 0 | `codex_clipboard_attachment`, `weak_evidence` | pass |
@@ -55,6 +56,7 @@ This benchmark runs the public fixture pack that ships with the repository and p
55
56
  | Codex file tree and workspace navigation UI failure | `fixtures/codex-file-tree-ui.md` | 75 | 2 | 0 | `codex_file_tree_ui`, `weak_evidence` | pass |
56
57
  | Codex session resume and state failure | `fixtures/codex-session-state.md` | 59 | 3 | 0 | `codex_resource_leak`, `codex_session_state`, `weak_evidence` | pass |
57
58
  | Codex usage bucket scope and percentage confusion | `fixtures/codex-usage-bucket-confusion.md` | 59 | 3 | 0 | `codex_token_burn`, `codex_usage_bucket_confusion`, `weak_evidence` | pass |
59
+ | Codex context or token usage indicator missing | `fixtures/codex-context-visibility.md` | 75 | 2 | 0 | `codex_context_visibility`, `weak_evidence` | pass |
58
60
  | Codex token burn and usage-drain loop | `fixtures/codex-token-burn.md` | 75 | 2 | 0 | `codex_token_burn`, `weak_evidence` | pass |
59
61
  | Codex resource leak and runaway process | `fixtures/codex-resource-leak.md` | 75 | 2 | 0 | `codex_resource_leak`, `weak_evidence` | pass |
60
62
  | Codex tool-call integrity and rollback failure | `fixtures/codex-tool-call-integrity.md` | 43 | 4 | 0 | `codex_resource_leak`, `codex_subagent_lifecycle`, `codex_tool_call_integrity`, `weak_evidence` | pass |
package/docs/USE_CASES.md CHANGED
@@ -59,7 +59,7 @@ What it proves:
59
59
  Recommended CI surface:
60
60
 
61
61
  ```yaml
62
- - uses: grnbtqdbyx-create/trace-to-skill@v0.1.94
62
+ - uses: grnbtqdbyx-create/trace-to-skill@v0.1.96
63
63
  with:
64
64
  mode: all
65
65
  doctor-threshold: "85"
@@ -198,6 +198,19 @@ This catches signals such as forked conversations carrying the full parent trans
198
198
 
199
199
  Include Codex app/CLI/extension version, surface, model, fork source thread id, forked thread id, fork action timestamp, fork boundary marker, `input_tokens` and `cached_input_tokens` before and after the fork, `prompt_cache_key` before and after, cache hit rate, duplicated parent-turn or tool-transcript examples with line ids, whether new files were read before the token jump, compaction state, subagent or `fork_context` history, minimal reproduction steps, and whether a fresh thread or non-fork continuation avoids the bloat.
200
200
 
201
+ ## 9.1. Codex Model Routing Mismatch Evidence
202
+
203
+ Use this when Codex shows one selected model in config, TUI, CLI flags, or UI, but server-side response evidence shows a different model was actually used.
204
+
205
+ ```bash
206
+ npx trace-to-skill demo model-routing-mismatch
207
+ npx trace-to-skill codex-report ./runs --output openai-codex-model-routing.md
208
+ ```
209
+
210
+ This catches signals such as `gpt-5.3-codex` being routed to `gpt-5.2`, `response.created` / `response.model` mismatch in SSE logs, `RUST_LOG=codex_api::sse::responses=trace` reproduction commands, silent fallback, no warning, and no fallback notice.
211
+
212
+ Include the Codex app/CLI/extension version, subscription/workspace, selected model from `config.toml`, TUI, command flag, or UI, actual server-side model from SSE `response.created` / `response.model`, the exact redacted `RUST_LOG` command, timestamp, account or verification state without secrets, whether API and Codex routes differ, whether a warning/fallback notice appeared, and a minimal one-prompt reproduction.
213
+
201
214
  ## 10. Codex Subagent Prompt Leakage Evidence
202
215
 
203
216
  Use this when `spawn_agent` child tasks are not isolated, especially with `fork_turns: "none"` or same-turn parallel subagent spawning.
@@ -322,6 +335,19 @@ This catches signals such as `codex resume` picker hangs, `codex resume <id>` wo
322
335
 
323
336
  For mixed resume, crash, config, plugin, or history issues, `diagnostics-bundle` writes the session, config, and plugin reports together with a checklist of files not to attach publicly.
324
337
 
338
+ ## 18.1. Codex Context Visibility Evidence
339
+
340
+ Use this when Codex Desktop no longer shows passive context/token usage visibility in the chat UI and long-session compaction pressure becomes invisible.
341
+
342
+ ```bash
343
+ npx trace-to-skill demo context-visibility
344
+ npx trace-to-skill codex-report ./runs --output openai-codex-context-visibility.md
345
+ ```
346
+
347
+ This catches signals such as a missing visible context/token usage indicator, hidden context-window pressure display, missing tooltip near the input area, `/status` not being a replacement for passive context awareness, local session logs still containing context data, and users being unable to tell when to compact, start a new thread, reduce pasted context, or split work before context loss.
348
+
349
+ Include Codex Desktop version, OS, surface, screenshot or short recording of the chat input area, whether the prior context/token indicator or tooltip was visible before the update, exact UI route where it disappeared, local session metadata showing context/window pressure if available, `/status` output if relevant, compaction timing, whether CLI/TUI still exposes a statusline, and how the missing indicator affects long-session decisions.
350
+
325
351
  ## 19. Codex File Tree UI Evidence
326
352
 
327
353
  Use this when Codex Desktop cannot reveal project files through the native file tree, folder icon, floating file panel, or built-in preview.
package/fixtures/codex-context-visibility.md ADDED
@@ -0,0 +1,18 @@
1
+ # Codex context visibility fixture
2
+
3
+ ## Public-style issue signals
4
+
5
+ - Codex Desktop no longer shows a visible context/token usage indicator in the chat UI after an update.
6
+ - Previously the app exposed context usage information, context-window pressure, or a tooltip near the input area.
7
+ - The current desktop app experience makes it difficult to understand how much context is being used, when compaction is likely, or whether a long-running session is approaching a practical context limit.
8
+ - `/status` is useful as an explicit command, but it is not a replacement for passive context awareness during long-running desktop threads.
9
+ - The missing indicator affects professional coding workflows because users cannot tell when to compact, start a new thread, reduce pasted context, or split work before context loss.
10
+ - A related report says the data exists in local session logs, but the app no longer exposes it passively.
11
+
12
+ ## Evidence checklist
13
+
14
+ - Codex Desktop version, OS, subscription, and surface.
15
+ - Screenshot or short recording of the chat input area where the context indicator or tooltip used to appear.
16
+ - Prior version where the context/token indicator was visible, current version where it is hidden or missing, and whether CLI/TUI still exposes a statusline.
17
+ - Local session metadata or `/status` output if available, with private paths and account details redacted.
18
+ - How the missing indicator affects long-session compaction, context-loss, or thread-splitting decisions.
package/fixtures/codex-model-routing-mismatch.md ADDED
@@ -0,0 +1,18 @@
1
+ # Codex model routing mismatch fixture
2
+
3
+ ## Public-style issue signals
4
+
5
+ - GPT-5.3-Codex is being routed to GPT-5.2.
6
+ - Both `config.toml` and the TUI are set to `gpt-5.3-codex`, but SSE captures show the actual `response.model` is `gpt-5.2-2025-12-11`.
7
+ - Running `RUST_LOG='codex_tui::chatwidget=info,codex_api::sse::responses=trace' codex` and sending a prompt shows `response.created` with `response.model=gpt-5.2-2025-12-11`.
8
+ - `codex exec --sandbox read-only --model gpt-5.3-codex 'ping'` logs the actual model as `gpt-5.2-2025-12-11`.
9
+ - The user sees no warning or fallback notice that a different model version is being used internally.
10
+ - Some reports mention ChatGPT Pro, WSL, macOS, recent CLI versions, and verification briefly restoring GPT-5.3-Codex before silently rerouting back to GPT-5.2.
11
+
12
+ ## Evidence checklist
13
+
14
+ - Codex app, CLI, or extension version and OS.
15
+ - Subscription/workspace and selected model from `config.toml`, TUI, command flag, or UI.
16
+ - Actual server-side model from SSE `response.created` / `response.model`.
17
+ - Exact redacted `RUST_LOG` or reproduction command, timestamp, and whether API and Codex routes differ.
18
+ - Whether a warning, fallback notice, account-verification message, or model-unavailable error appeared.
package/fixtures/github-codex-issues-export.json CHANGED
@@ -61,6 +61,88 @@
61
61
  "reactions": { "totalCount": 8 },
62
62
  "updatedAt": "2025-10-16T20:32:11Z"
63
63
  },
64
+ {
65
+ "number": 11189,
66
+ "title": "GPT-5.3-Codex being routed to GPT-5.2",
67
+ "body": "Both config.toml and TUI are set for gpt-5.3-codex, but output and SSE captures show response.model is actually gpt-5.2-2025-12-11. The user selected GPT-5.3-Codex but is silently rerouted to GPT-5.2 with no fallback notice.",
68
+ "url": "https://github.com/openai/codex/issues/11189",
69
+ "labels": [
70
+ { "name": "bug" },
71
+ { "name": "CLI" }
72
+ ],
73
+ "commentsCount": 169,
74
+ "reactions": { "totalCount": 12 },
75
+ "updatedAt": "2026-05-05T22:11:18Z"
76
+ },
77
+ {
78
+ "number": 11561,
79
+ "title": "GPT-5.3-Codex being routed to GPT-5.2",
80
+ "body": "Set both config.toml and TUI to gpt-5.3-codex, run RUST_LOG=codex_tui::chatwidget=info,codex_api::sse::responses=trace codex, send a prompt, and log/codex-tui.log shows response.model is gpt-5.2-2025-12-11.",
81
+ "url": "https://github.com/openai/codex/issues/11561",
82
+ "labels": [
83
+ { "name": "bug" },
84
+ { "name": "CLI" }
85
+ ],
86
+ "commentsCount": 47,
87
+ "reactions": { "totalCount": 4 },
88
+ "updatedAt": "2026-03-03T15:08:52Z"
89
+ },
90
+ {
91
+ "number": 11842,
92
+ "title": "GPT-5.3-Codex being routed to GPT-5.2",
93
+ "body": "Although the CLI displays gpt-5.3-codex, inspecting the actual SSE response stream shows that the request is routed to gpt-5.2. There is no warning or fallback notice indicating a different model version is used internally.",
94
+ "url": "https://github.com/openai/codex/issues/11842",
95
+ "labels": [
96
+ { "name": "bug" },
97
+ { "name": "CLI" }
98
+ ],
99
+ "commentsCount": 15,
100
+ "reactions": { "totalCount": 2 },
101
+ "updatedAt": "2026-02-18T18:34:33Z"
102
+ },
103
+ {
104
+ "number": 23794,
105
+ "title": "Codex Desktop no longer shows visible context/token usage indicator",
106
+ "body": "After updating Codex Desktop, the app no longer shows a visible context/token usage indicator in the chat UI. Previously Codex exposed context-window pressure near the input area or usage indicator. Without it users cannot tell when a thread is close to compaction, context pressure, or context loss.",
107
+ "url": "https://github.com/openai/codex/issues/23794",
108
+ "labels": [
109
+ { "name": "bug" },
110
+ { "name": "context" },
111
+ { "name": "app" }
112
+ ],
113
+ "commentsCount": 160,
114
+ "reactions": { "totalCount": 20 },
115
+ "updatedAt": "2026-06-01T01:17:38Z"
116
+ },
117
+ {
118
+ "number": 23591,
119
+ "title": "Reimplement visible context/token usage indicator in Codex Desktop App",
120
+ "body": "Request to restore or re-add a visible context/token usage indicator on mouse-over. The desktop app makes it difficult to understand how much context is being used, when compaction is likely, and whether a long-running session is approaching a practical context limit.",
121
+ "url": "https://github.com/openai/codex/issues/23591",
122
+ "labels": [
123
+ { "name": "enhancement" },
124
+ { "name": "rate-limits" },
125
+ { "name": "context" },
126
+ { "name": "app" }
127
+ ],
128
+ "commentsCount": 7,
129
+ "reactions": { "totalCount": 5 },
130
+ "updatedAt": "2026-05-29T23:12:20Z"
131
+ },
132
+ {
133
+ "number": 24710,
134
+ "title": "Codex Desktop: hidden context indicator still blocks long-session context management",
135
+ "body": "The Codex Desktop app no longer shows an always-visible context/token usage indicator after the update. The data exists in local session logs, but the app no longer exposes it passively, which blocks long-session context management.",
136
+ "url": "https://github.com/openai/codex/issues/24710",
137
+ "labels": [
138
+ { "name": "enhancement" },
139
+ { "name": "context" },
140
+ { "name": "app" }
141
+ ],
142
+ "commentsCount": 1,
143
+ "reactions": { "totalCount": 1 },
144
+ "updatedAt": "2026-05-27T05:08:25Z"
145
+ },
64
146
  {
65
147
  "number": 6465,
66
148
  "title": "MCP servers not detected in Codex VS Code extension but working in Codex CLI",
package/llms.txt CHANGED
@@ -82,6 +82,7 @@ failed agent run -> failure class -> evidence-backed AGENTS.md/SKILL.md suggesti
82
82
  ```bash
83
83
  npx trace-to-skill doctor .
84
84
  npx trace-to-skill demo
85
+ npx trace-to-skill demo model-routing-mismatch
85
86
  npx trace-to-skill demo thinking-hang
86
87
  npx trace-to-skill demo clipboard-attachment
87
88
  npx trace-to-skill demo deeplink-launch
@@ -90,6 +91,7 @@ npx trace-to-skill demo auth-verification
90
91
  npx trace-to-skill demo context-fork-bloat
91
92
  npx trace-to-skill demo subagent-prompt-leakage
92
93
  npx trace-to-skill demo usage-bucket-confusion
94
+ npx trace-to-skill demo context-visibility
93
95
  npx trace-to-skill demo mcp-discovery-mismatch
94
96
  npx trace-to-skill demo mcp-streamable-http
95
97
  npx trace-to-skill demo hooks-runtime
@@ -129,7 +131,7 @@ gh issue list --repo openai/codex --state all --limit 100 --json number,title,bo
129
131
  ## GitHub Action
130
132
 
131
133
  ```yaml
132
- - uses: grnbtqdbyx-create/trace-to-skill@v0.1.94
134
+ - uses: grnbtqdbyx-create/trace-to-skill@v0.1.96
133
135
  with:
134
136
  mode: all
135
137
  doctor-threshold: "85"
@@ -142,7 +144,7 @@ gh issue list --repo openai/codex --state all --limit 100 --json number,title,bo
142
144
  ## Weekly Codex Issue Radar
143
145
 
144
146
  ```yaml
145
- - uses: grnbtqdbyx-create/trace-to-skill@v0.1.94
147
+ - uses: grnbtqdbyx-create/trace-to-skill@v0.1.96
146
148
  with:
147
149
  mode: issue-map
148
150
  issue-map-repo: openai/codex
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "trace-to-skill",
3
- "version": "0.1.94",
3
+ "version": "0.1.96",
4
4
  "description": "Turn failed AI coding-agent runs into reusable AGENTS.md rules, SKILL.md files, and eval evidence.",
5
5
  "type": "module",
6
6
  "main": "dist/src/index.js",
@@ -68,6 +68,11 @@
68
68
  "context-compaction",
69
69
  "codex-compaction",
70
70
  "codex-context-drift",
71
+ "codex-model-routing",
72
+ "codex-model-mismatch",
73
+ "codex-model-fallback",
74
+ "gpt-5-3-codex",
75
+ "gpt-5-2-codex",
71
76
  "latest-message",
72
77
  "codex-latency",
73
78
  "codex-thinking-hang",
@@ -164,6 +169,10 @@
164
169
  "codex-token-burn",
165
170
  "codex-usage",
166
171
  "codex-usage-bucket",
172
+ "codex-context-visibility",
173
+ "context-token-indicator",
174
+ "context-usage-indicator",
175
+ "context-window-pressure",
167
176
  "usage-popover",
168
177
  "rate-limit-ui",
169
178
  "codex-reset",