fix(ui): show only Standard and Fast for OpenAI speed by steipete · Pull Request #100190 · openclaw/openclaw

steipete · 2026-07-05T06:16:05Z

What Problem This Solves

Fixes an issue where OpenAI users opening the Control UI model picker would see four speed choices—Default, Fast, Standard, and Auto—even though OpenAI exposes only the standard and priority service tiers.

Why This Change Was Made

The OpenAI picker now presents only Standard and Fast, ordered Standard first. Selection follows the effective session state so inherited Fast remains visible; legacy Auto remains unselected until the user explicitly chooses one of the supported binary modes. Other providers retain their existing choices.

User Impact

OpenAI users now get an unambiguous two-choice speed control matching the provider contract.

Evidence

Reproduced the four-choice OpenAI picker in the running Control UI.
OPENCLAW_HEAVY_CHECK_LOCK_SCOPE=worktree OPENCLAW_VITEST_FS_MODULE_CACHE_PATH=/tmp/openclaw-speed-picker-vitest node scripts/run-vitest.mjs ui/src/lib/chat/model-select-state.test.ts — 11 tests passed.
OPENCLAW_HEAVY_CHECK_LOCK_SCOPE=worktree node scripts/run-tsgo.mjs -p tsconfig.core.json — passed.
.agents/skills/autoreview/scripts/autoreview --mode local --stream-engine-output — clean after addressing effective/inherited mode selection.
Verified the dependency contract in sibling Codex sources: codex-rs/protocol/src/config_types.rs, codex-rs/protocol/src/openai_models.rs, codex-rs/core/src/session/mod.rs, and their service-tier tests. Codex maps Standard to the default tier and Fast to priority.

AI-assisted: implemented and reviewed with Codex; maintainer-directed change.

clawsweeper · 2026-07-05T06:18:02Z

Codex review: needs real behavior proof before merge. Reviewed July 5, 2026, 4:16 AM ET / 08:16 UTC.

Summary
The PR changes the Control UI chat speed-state resolver and tests so OpenAI models show a Standard/Fast picker, plus a changelog entry.

PR surface: Source +16, Tests +62, Docs +1. Total +79 across 3 files.

Reproducibility: yes. Current main source shows OpenAI receives the generic four-option speed picker, and PR-head source shows the inherited/default-state regressions without needing a live UI run.

Review metrics: 1 noteworthy metric.

OpenAI speed choices: 4 generic choices narrowed to 2 OpenAI choices. The semantic reduction changes how users persist or clear session fast-mode overrides, so compatibility needs review before merge.

Merge readiness
Overall: 🧂 unranked krab
Proof: 🧂 unranked krab
Patch quality: 🧂 unranked krab
Result: blocked until real behavior proof is added.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

[P1] Fix the inherited/default speed-state path so selected inherited clicks do not persist overrides and stored overrides can be cleared.
[P1] Add redacted after-fix Control UI proof, preferably a screenshot or short recording that shows the OpenAI picker and the clear/inherit behavior.

Proof guidance:

[P1] Needs real behavior proof before merge: The PR body has tests, CI, and a before-fix observation, but no after-fix screenshot, recording, terminal output, live output, linked artifact, or redacted logs showing the changed Control UI picker. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.

Mantis proof suggestion
A short visual Control UI proof would materially show the OpenAI speed picker and inherited/default behavior that unit tests alone do not demonstrate. A maintainer can ask Mantis to capture proof by posting this exact PR comment:

@openclaw-mantis visual task: verify the Control UI OpenAI speed picker shows Standard/Fast and users can clear or inherit speed without selected inherited clicks pinning overrides.

Risk before merge

[P1] Merging as-is can leave OpenAI users without a picker path to clear a stored session speed override and return to inherited/default speed.
[P1] Merging as-is can turn inherited OpenAI Fast or Standard state into an explicit session override when the already selected option is clicked.
[P1] The PR still has no after-fix real Control UI proof showing the changed picker and inheritance/clearing behavior.

Maintainer options:

Restore inherited/default speed semantics (recommended)
Keep the two OpenAI service-tier choices while adding a clear/inherit path and preventing selected inherited clicks from persisting session overrides.
Accept binary-only persistence
Maintainers could intentionally accept that the OpenAI picker becomes binary-only, but that should be an explicit compatibility/product choice.
Hold for visual proof
Pause merge until after-fix Control UI proof shows the OpenAI picker and the inherited/default recovery path.

Next step before merge

[P1] The maintainer-labeled PR has concrete UI state defects plus missing after-fix real behavior proof, so it needs author or maintainer follow-up rather than cleanup closure or automated repair.

Security
Cleared: The diff only changes Control UI resolver/test logic and a changelog line, with no new dependency, workflow, secret, permission, or code-execution surface.

Review findings

[P1] Preserve inherited speed on selected clicks — ui/src/lib/chat/model-select-state.ts:205-210
[P1] Keep a Default path for OpenAI speed overrides — ui/src/lib/chat/model-select-state.ts:225-229

Review details

Best possible solution:

Keep the OpenAI Standard/Fast simplification, but preserve an inherit/clear state for stored overrides and make selected inherited choices no-op, then add real Control UI proof.

Do we have a high-confidence way to reproduce the issue?

Yes. Current main source shows OpenAI receives the generic four-option speed picker, and PR-head source shows the inherited/default-state regressions without needing a live UI run.

Is this the best way to solve the issue?

No. Narrowing OpenAI to Standard/Fast matches the Codex service-tier contract, but this patch is not the best final fix until it preserves clearing and inherited selected-state behavior.

Full review comments:

[P1] Preserve inherited speed on selected clicks — ui/src/lib/chat/model-select-state.ts:205-210
OpenAI rows with no stored fastMode can now render inherited Fast or Standard as selected, but clicking that selected option still flows through switchChatFastMode, which only no-ops against stored activeRow.fastMode. That writes a session override instead of preserving inheritance.
Confidence: 0.92
[P1] Keep a Default path for OpenAI speed overrides — ui/src/lib/chat/model-select-state.ts:225-229
The OpenAI option list no longer includes the empty value that switchChatFastMode uses to send fastMode: null and clear an override. Choosing Standard stores false, so users cannot return an OpenAI chat to inherited config/agent/model speed from this picker. Late discovery: this was already visible in earlier reviewed heads.
Confidence: 0.9
Late finding: first raised on code an earlier review cycle already covered.

Overall correctness: patch is incorrect
Overall confidence: 0.9

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 07bf384a8b3f.

Label changes

Label justifications:

P2: The PR addresses a bounded Control UI bug, but the remaining defects affect persisted session speed settings.
merge-risk: 🚨 compatibility: The diff can remove the OpenAI clear/inherit affordance and pin inherited fast-mode settings as session overrides.
rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🧂 unranked krab.
status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask. Needs real behavior proof before merge: The PR body has tests, CI, and a before-fix observation, but no after-fix screenshot, recording, terminal output, live output, linked artifact, or redacted logs showing the changed Control UI picker. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.

Evidence reviewed

PR surface:

Source +16, Tests +62, Docs +1. Total +79 across 3 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	1	24	8	+16
Tests	1	62	0	+62
Docs	1	1	0	+1
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	3	87	8	+79

What I checked:

PR-head OpenAI picker options: At head 58d32da, OpenAI uses only off/on speed options, removing the empty Default option from the provider-specific picker. (ui/src/lib/chat/model-select-state.ts:225, 58d32da27bfa)
PR-head inherited selection: OpenAI currentOverride is derived from effectiveFastMode ?? fastMode, so inherited Fast or Standard can render as selected without a stored session override. (ui/src/lib/chat/model-select-state.ts:205, 58d32da27bfa)
Speed option click path: The rendered speed button still dispatches onFastModeSelect(speedValue) for the selected option, so selected inherited values are sent through the mutation path. (ui/src/pages/chat/components/chat-model-controls.ts:376, 58d32da27bfa)
Stored-only no-op and clear contract: switchChatFastMode compares only stored activeRow.fastMode, maps the empty value to undefined, and sends null to clear; choosing Standard sends false instead. (ui/src/pages/chat/chat-session.ts:256, 58d32da27bfa)
Gateway clearing behavior: The gateway deletes stored fastMode when the patch value is null, confirming the empty picker value is the current UI clear/inherit route. (src/gateway/sessions-patch.ts:413, 07bf384a8b3f)
Effective fast mode exists in session rows: Current main exposes both stored fastMode and inherited effectiveFastMode, so inherited speed state is a real runtime shape the picker must preserve. (src/gateway/session-utils.ts:2181, 07bf384a8b3f)

Likely related people:

shakkernerd: Merged Control UI architecture work added the current model-select-state, chat controls, and chat session modules that now own the speed picker flow. (role: introduced current Control UI chat settings surface; confidence: high; commits: 65e12328aa20; files: ui/src/lib/chat/model-select-state.ts, ui/src/pages/chat/components/chat-model-controls.ts, ui/src/pages/chat/chat-session.ts)
steipete: The PR is assigned to steipete and current path history shows prior fast-mode/runtime work in src/agents/fast-mode.ts and session-related changes. (role: current assignee and fast-mode adjacent contributor; confidence: medium; commits: dddf871ad969, 6df0fb818d67, 58d32da27bfa; files: src/agents/fast-mode.ts, src/gateway/session-utils.ts, ui/src/lib/chat/model-select-state.ts)
vincentkoc: Recent history on fast-mode resolution includes the auto-mode forward-port that feeds the effective fast-mode state consumed by this UI. (role: fast-mode feature contributor; confidence: medium; commits: 2b75806197ab; files: src/agents/fast-mode.ts, src/gateway/session-utils.ts)

What the crustacean ranks mean

🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

Review history (7 earlier review cycles)

reviewed 2026-07-05T06:21:44.940Z sha e4c69c8 :: needs real behavior proof before merge. :: none
reviewed 2026-07-05T06:45:53.418Z sha d184bf5 :: needs real behavior proof before merge. :: [P2] Avoid pinning inherited Fast when the selected option is clicked
reviewed 2026-07-05T06:54:22.383Z sha d184bf5 :: needs real behavior proof before merge. :: [P2] Preserve inherited speed state on selected clicks
reviewed 2026-07-05T07:13:01.751Z sha 63576ba :: needs real behavior proof before merge. :: [P2] Preserve inherited speed state on selected clicks
reviewed 2026-07-05T07:20:25.708Z sha 63576ba :: needs real behavior proof before merge. :: [P2] Preserve inherited speed state on selected clicks
reviewed 2026-07-05T07:37:33.888Z sha 687db25 :: needs real behavior proof before merge. :: [P2] Preserve inherited speed on selected clicks
reviewed 2026-07-05T07:45:04.906Z sha 687db25 :: needs real behavior proof before merge. :: [P2] Preserve inherited speed on selected clicks

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e4c69c8a40

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-07-05T06:19:57Z

          ? "off"
          : "";
+  const isOpenAI = effectiveProvider === "openai";
+  const effectiveOpenAIMode = activeRow?.effectiveFastMode ?? activeRow?.fastMode;


Preserve inherited speed state on selected clicks

When an OpenAI row inherits speed from config/agent/default (fastMode is undefined but effectiveFastMode is true/false, which sessions.list emits), this makes Fast/Standard render as the selected option. The click path still no-ops only when previousFastMode === next in switchChatFastMode, so clicking the already-selected choice sends sessions.patch with a session-level fastMode override and breaks future inheritance. Either keep inherited state distinct or make the handler compare against the effective mode before patching.

Useful? React with 👍 / 👎.

steipete · 2026-07-05T06:40:10Z

Land-ready on exact head e4c69c8a40dba463a6a3c20dea487f658194eab7.

Proof:

OPENCLAW_HEAVY_CHECK_LOCK_SCOPE=worktree OPENCLAW_VITEST_FS_MODULE_CACHE_PATH=/tmp/openclaw-speed-picker-vitest node scripts/run-vitest.mjs ui/src/lib/chat/model-select-state.test.ts — 11 passed.
OPENCLAW_HEAVY_CHECK_LOCK_SCOPE=worktree node scripts/run-tsgo.mjs -p tsconfig.core.json — passed.
Fresh Codex autoreview — clean after correcting inherited/effective Fast selection.
Exact-head CI run 28731728643 attempt 2 — success. Initial QA aggregate and a broad Vitest shard were transient; all 83 QA scenarios passed on attempt 1, and both jobs passed on rerun.
scripts/pr review-validate-artifacts 100190 and scripts/pr prepare-run 100190 — passed.

OpenAI now exposes only Standard and Fast in the picker. Other providers retain their existing choices; inherited Fast and legacy Auto normalization are covered.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d184bf5d41

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-07-05T06:45:33Z

 const COMMAND_ALIASES: Record<string, string> = {
-  elev: "elevated",
  gwstatus: "gateway-status",
 };


Restore the /elev alias mapping

When a TUI user enters the still-advertised /elev full shortcut, parseCommand now falls through to COMMAND_ALIASES after resolveTextCommand finds no registry alias for /elev, but this map no longer rewrites elev to elevated. The command handler only has a case for elevated, so /elev ... is treated as an arbitrary slash message and sent to the model instead of patching the elevated mode.

Useful? React with 👍 / 👎.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 58d32da27b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-07-05T08:07:14Z

+      ? [
+          { value: "off", label: "Standard" },
+          { value: "on", label: "Fast" },
+        ]


Preserve a way to clear OpenAI speed overrides

When an OpenAI session already has a session-level speed override, this OpenAI-only option list no longer includes the empty Default value. The picker’s handler only clears an override when switchChatFastMode receives ""; choosing Standard sends fastMode: false, so users cannot return an OpenAI chat to model/agent/config inheritance from the picker after choosing Fast or Standard. Keep a clear/default affordance or otherwise route the intended inherited state to the empty value.

Useful? React with 👍 / 👎.

steipete · 2026-07-05T08:47:55Z

Merged via squash.

Prepared head SHA: 217521a2710cf972e1201dbac6dd63abef7e6449
Landed commit: 8b229a938c28198e163c821a8fce13ac88818e33

openclaw-barnacle Bot added app: web-ui App: web-ui size: S maintainer Maintainer-authored PR labels Jul 5, 2026

steipete self-assigned this Jul 5, 2026

chatgpt-codex-connector Bot reviewed Jul 5, 2026

View reviewed changes

openclaw-barnacle Bot added docs Improvements or additions to documentation channel: whatsapp-web Channel integration: whatsapp-web app: ios App: ios app: macos App: macos scripts Repository scripts extensions: qa-lab plugin: google-meet size: XL and removed size: S labels Jul 5, 2026

chatgpt-codex-connector Bot reviewed Jul 5, 2026

View reviewed changes

clawsweeper Bot added the status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. label Jul 5, 2026

openclaw-barnacle Bot added commands Command implementations agents Agent runtime and tooling labels Jul 5, 2026

steipete force-pushed the codex/openai-speed-picker branch from 63576ba to 687db25 Compare July 5, 2026 07:32

clawsweeper Bot added rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. and removed rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. labels Jul 5, 2026

steipete force-pushed the codex/openai-speed-picker branch from 687db25 to 58d32da Compare July 5, 2026 08:03

chatgpt-codex-connector Bot reviewed Jul 5, 2026

View reviewed changes

steipete force-pushed the codex/openai-speed-picker branch from 58d32da to 59a1b36 Compare July 5, 2026 08:18

fix(ui): simplify OpenAI speed choices

217521a

steipete force-pushed the codex/openai-speed-picker branch from 59a1b36 to 217521a Compare July 5, 2026 08:28

steipete merged commit 8b229a9 into main Jul 5, 2026
124 of 125 checks passed

steipete deleted the codex/openai-speed-picker branch July 5, 2026 08:46

github-actions Bot mentioned this pull request Jul 5, 2026

📡 Upstream Digest — 2026-07-05 10:03 UTC curtismercier/openclaw-mods#1212

Open

github-actions Bot pushed a commit to Desicool/openclaw that referenced this pull request Jul 6, 2026

fix(ui): simplify OpenAI speed choices (openclaw#100190)

184170b

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(ui): show only Standard and Fast for OpenAI speed#100190

fix(ui): show only Standard and Fast for OpenAI speed#100190
steipete merged 1 commit into
mainfrom
codex/openai-speed-picker

steipete commented Jul 5, 2026

Uh oh!

clawsweeper Bot commented Jul 5, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Uh oh!

steipete commented Jul 5, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Uh oh!

Uh oh!

steipete commented Jul 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Uh oh!

Conversation

steipete commented Jul 5, 2026

What Problem This Solves

Why This Change Was Made

User Impact

Evidence

Uh oh!

clawsweeper Bot commented Jul 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Choose a reason for hiding this comment

Uh oh!

steipete commented Jul 5, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jul 5, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

steipete commented Jul 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

clawsweeper Bot commented Jul 5, 2026 •

edited

Loading