feat(orchestrator): Surface Grok reasoning effort options

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.

[mwolson]

Summary

• Add a Reasoning selector for Grok models that advertise effort support, using
  each model's ACP menu (Grok 4.5: Low / Medium / High; Grok 4.6 also includes
  Extra High).
• Require Grok 1.0.0 or newer and apply model and reasoning changes in the
  active conversation through ACP session/set_model metadata.
• Keep the active model and reasoning metadata synchronized across new, loaded,
  resumed, and restarted ACP sessions, including first-party text generation.
• When a model advertises more than one default, keep a single Default badge and
  prefer the advertised current effort.

Problem and Fix

Problem and Why it Happened	Fix
T3 Code did not surface Grok's advertised reasoning-effort choices, so every turn used the provider default.	Map ACP model reasoning metadata into the existing provider-option UI on web and mobile.
Grok 1.0.0 honors reasoning effort in session/set_model metadata, while the generic ACP adapter previously treated model selection as only a model-id change.	Compare both the model id and normalized { reasoningEffort } metadata so the next turn can apply either change in the same conversation.
Loading or resuming a provider session can replace the model state discovered when the process initialized.	Record the active model and metadata after every session activation, then reapply the requested selection when necessary.
An older Grok binary can accept process launch but ignore the dynamic reasoning metadata.	Parse the complete Grok CLI version, including prerelease suffixes, and gate orchestration and first-party text generation on Grok 1.0.0 or newer.
Grok 4.6's catalog marks Extra High and High as default: true, so the picker showed two Default badges.	Collapse multiple advertised defaults to one and prefer the advertised current effort.

Defensive Fixes

Problem and Why it Happened	Fix
Provider metadata can be absent or malformed, and persisted selections can drift from a model's current advertised menu.	Fall back only when the effort menu is absent, reject malformed-present metadata, preserve future valid advertised values, and normalize stale values to the model's advertised default.
Clearing reasoning metadata is also a state change, but a one-sided comparison can miss the transition from a defined effort to no explicit effort.	Compare the current and requested metadata symmetrically and cover clearing and reapplying effort on the same model.

UI Changes

After: Grok models with reasoning support expose their advertised choices in
the composer. Grok 4.6 shows Extra High when that is the advertised current
effort, with a single Default badge in the open menu.

A video is not applicable because this change has no animation or timing
behavior.

Validation

• A live Grok ACP probe changed an active session to Low, completed a real
  prompt, loaded the same session in a fresh runtime, and read Low back from
  the advertised model metadata. Grok's durable session history recorded
  "reasoning_effort":"low" on those assistant messages.
• A later live Grok 4.6 thread changed mid-conversation from High to Extra
  High. Subsequent assistant messages recorded "reasoning_effort":"xhigh".
• After collapsing duplicate advertised defaults, a Nightly AppImage on
  Grok 4.6 shows Extra High selected and a single Default badge on Extra High.
• Focused Grok provider tests cover advertised menus, the Grok 4.6 dual-default
  catalog, and Extra High as the current effort.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included after screenshots for UI changes
• A video is not applicable because there are no animation or timing changes

Note

Surface Grok reasoning effort as a spawn-bound option in the orchestrator and UI

• Adds GROK_REASONING_EFFORT_OPTION_ID and supporting utilities in GrokAcpSupport.ts to resolve and pass --reasoning-effort as a CLI flag when spawning the Grok agent process.
• Extends the ACP adapter in AcpAdapterV2.ts to support spawnOptionIds on flavors, excluding those options from ACP config validation and setConfigOption calls, and logging a warning on mismatch.
• Updates GrokProvider.ts to parse reasoning effort menus from ACP model _meta and advertise a Reasoning select capability on discovered Grok models, with a fallback for grok-4.5.
• Blocks option changes mid-thread on both web (ChatComposer.tsx) and mobile (ThreadComposer.tsx) when the active provider requires a new thread for model changes, showing a warning toast and reverting to the committed selection.
• Behavioral Change: reasoning effort for Grok is now applied at spawn time rather than via session config, so changing it mid-thread is blocked and requires starting a new thread.

📊 Macroscope summarized 64cbc0c. 26 files reviewed, 0 issues evaluated, 0 issues filtered, 0 comments posted

🗂️ Filtered Issues

No issues evaluated.

>

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 72a67341-bf75-4496-ac58-dd45f02f89ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[mwolson]

After: Grok 4.6 reasoning selector

Composer after the change. Grok 4.6 is selected with Extra High as the current reasoning effort.

[mwolson]

After: Grok 4.6 reasoning menu

Open Reasoning menu. Extra High is the only Default.

Composer with Grok 4.6 and the open Extra High menu.

[macroscopeapp]

Reviewed the web UI changes for the session-bound provider-option lock (ChatView, ChatComposer, TraitsPicker, composerProviderState). Prop forwarding through composerProviderState and the display/dispatch split look consistent with the shared control contracts. Two findings on the new locked-option interaction.

^{Posted via Macroscope — UI Consistency}

[t3dotgg]

Note

🤖 GPT-5.6 Sol responding on behalf of Theo

Closing this PR after an automated pass over open pull requests. Large stale reasoning-controls draft duplicates trusted Grok work in #6386.