Add ACP support with Cursor provider

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}

[juliusmarminge]

Summary

• Adds Cursor as a first-class provider with ACP session lifecycle support, health checks, and adapter wiring in the server.
• Implements Cursor model selection, including fast/plan mode mapping and session restart behavior when model options change.
• Preserves provider/thread model state through orchestration, projection, and turn dispatch paths.
• Updates the web app to surface Cursor traits, provider/model selection, and session drafting behavior.
• Expands runtime ingestion so completed tool events retain structured tool metadata.

Testing

• bun fmt
• bun lint
• bun typecheck
• Added and updated tests across server, contracts, shared, and web layers for Cursor adapter behavior, orchestration routing, session model changes, and UI state handling.
• Not run: bun run test

---

Note

High Risk
High risk because it introduces a new Cursor ACP provider/agent integration and significantly changes orchestration/runtime-ingestion behavior around turn starts, tool/approval boundaries, and assistant message buffering/segmentation.

Overview
Adds Cursor as a first-class provider using ACP (stdio JSON-RPC), including a new CursorTextGenerationLive path for git text generation that spawns an ACP runtime, applies ACP-configured model options, and tolerates noisy JSON responses.

Extends server orchestration to route text generation to Cursor and refines provider turn start handling to better surface/record start failures without crashing. Runtime ingestion is reworked to segment assistant messages across approval/user-input boundaries, flush buffered text deterministically, avoid whitespace-only artifacts/duplicate completions, and preserve structured tool-call metadata for completed tool activities.

Also updates Claude model-id resolution to use provider-specific resolveClaudeApiModelId, raises the checkpoint diff git output cap, and adds effect-acp wiring/build config so ACP code can be bundled and tested.

^{Reviewed by Cursor Bugbot for commit aa696b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Add Cursor as a provider with ACP-backed sessions, text generation, and UI support

• Introduces a new effect-acp workspace package implementing a typed JSON-RPC client/agent transport over stdio, matching the ACP protocol (v0.11.3), with schema-generated types and Effect-native error handling.
• Adds CursorAdapterLive and CursorProviderLive server layers that launch the Cursor ACP agent subprocess, manage session lifecycle (start, send, interrupt, approve, stop), discover models/capabilities, and stream provider snapshots.
• Extends contracts (ProviderKind, ModelSelection, ProviderModelOptions, ServerSettings) and shared utilities to include Cursor-specific types: CursorModelOptions, CursorModelSelection, CursorSettings, and CursorSettingsPatch.
• Wires Cursor into the composer UI: model picker, TraitsPicker (fast mode, reasoning, context window), settings panel with 'Early Access' badge, and composerProviderRegistry.
• Adds CursorTextGenerationLive for ACP-backed git text generation (commit messages, PR content, branch names, thread titles) with a 180s timeout and JSON extraction.
• Improves ProviderRuntimeIngestion with segment-aware assistant message streaming, buffered flush on pause events, and suppression of empty completions.
• Improves ProviderCommandReactor turn-start failure handling: sets lastError on the thread session, appends a provider.turn.start.failed activity, and forks the send with recovery logging.
• Risk: ProviderSessionModelSwitchMode removes the 'restart-session' value; any persisted or in-flight state using that string will no longer match the type union.

^{Macroscope summarized aa696b5.}

[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

Run ID: c2d0780d-752d-49d8-b364-1135b6096558

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch t3code/greeting

---

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

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Merge never clears cached provider model options
  • Replaced ?? fallback with key in incoming check so explicitly-present-but-undefined provider keys now clear cached values, and added cache deletion when merge produces undefined.
• ✅ Fixed: Mock agent test uses strict equal with extra fields
  • Changed toEqual to toMatchObject so the assertion tolerates the extra modes field returned by the mock agent.

Or push these changes by commenting:

@cursor push beb68c40d7

Preview (beb68c40d7)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -50,20 +50,17 @@
cached: ProviderModelOptions | undefined,
incoming: ProviderModelOptions | undefined,
): ProviderModelOptions | undefined {
- if (!cached && !incoming) {- return undefined;+ if (incoming === undefined) return cached;+ if (cached === undefined) return incoming;++ const providerKeys = ["codex", "claudeAgent", "cursor"] as const;+ const next: Record<string, unknown> = {};+ for (const key of providerKeys) {+ const value = key in incoming ? incoming[key] : cached[key];+ if (value !== undefined) {+ next[key] = value;+ }
}
- const next = {- ...(incoming?.codex !== undefined || cached?.codex !== undefined- ? { codex: incoming?.codex ?? cached?.codex }- : {}),- ...(incoming?.claudeAgent !== undefined || cached?.claudeAgent !== undefined- ? { claudeAgent: incoming?.claudeAgent ?? cached?.claudeAgent }- : {}),- ...(incoming?.cursor !== undefined || cached?.cursor !== undefined- ? { cursor: incoming?.cursor ?? cached?.cursor }- : {}),- } satisfies Partial<ProviderModelOptions>;
return Object.keys(next).length > 0 ? (next as ProviderModelOptions) : undefined;
}
@@ -405,8 +402,12 @@
threadModelOptions.get(input.threadId),
input.modelOptions,
);
- if (mergedModelOptions !== undefined) {- threadModelOptions.set(input.threadId, mergedModelOptions);+ if (input.modelOptions !== undefined) {+ if (mergedModelOptions !== undefined) {+ threadModelOptions.set(input.threadId, mergedModelOptions);+ } else {+ threadModelOptions.delete(input.threadId);+ }
}
const normalizedInput = toNonEmptyProviderInput(input.messageText);
const normalizedAttachments = input.attachments ?? [];
diff --git a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts--- a/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts+++ b/apps/server/src/provider/acp/AcpJsonRpcConnection.test.ts@@ -32,7 +32,7 @@
cwd: process.cwd(),
mcpServers: [],
});
- expect(newResult).toEqual({ sessionId: "mock-session-1" });+ expect(newResult).toMatchObject({ sessionId: "mock-session-1" });
const promptResult = yield* conn.request("session/prompt", {
sessionId: "mock-session-1",

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

Autofix Details

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Redundant Schema.is check after findProviderAdapterRequestError
  • Replaced the redundant Schema.is(ProviderAdapterRequestError) checks with simple truthy checks since findProviderAdapterRequestError already validates and returns the typed result.
• ✅ Fixed: formatFailureDetail uses .message instead of .detail
  • Changed providerError.message to providerError.detail in formatFailureDetail to produce the specific human-readable error description consistent with the rest of the file.

Or push these changes by commenting:

@cursor push 0416bda917

Preview (0416bda917)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -98,7 +98,7 @@
function isUnknownPendingApprovalRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
const detail = error.detail.toLowerCase();
return (
detail.includes("unknown pending approval request") ||
@@ -114,7 +114,7 @@
function isUnknownPendingUserInputRequestError(cause: Cause.Cause<ProviderServiceError>): boolean {
const error = findProviderAdapterRequestError(cause);
- if (Schema.is(ProviderAdapterRequestError)(error)) {+ if (error) {
return error.detail.toLowerCase().includes("unknown pending user-input request");
}
return Cause.pretty(cause).toLowerCase().includes("unknown pending user-input request");
@@ -211,7 +211,7 @@
? failReason.error
: undefined;
if (providerError) {
- return providerError.message;+ return providerError.detail;
}
return Cause.pretty(cause);
};

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Stale model override after session restart for unsupported switch
  • When sessionModelSwitch is 'unsupported' and the user explicitly requested a model change (input.modelSelection is defined), the fix now passes through the user's requested model directly instead of overriding it with the stale activeSession.model, while preserving the session-model fallback for subsequent turns without an explicit model change.

Or push these changes by commenting:

@cursor push 245526b0fa

Preview (245526b0fa)

diff --git a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts--- a/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts+++ b/apps/server/src/orchestration/Layers/ProviderCommandReactor.ts@@ -428,12 +428,14 @@
input.modelSelection ?? threadModelSelections.get(input.threadId) ?? thread.modelSelection;
const modelForTurn =
sessionModelSwitch === "unsupported"
- ? activeSession?.model !== undefined- ? {- ...requestedModelSelection,- model: activeSession.model,- }- : requestedModelSelection+ ? input.modelSelection !== undefined+ ? input.modelSelection+ : activeSession?.model !== undefined+ ? {+ ...requestedModelSelection,+ model: activeSession.model,+ }+ : requestedModelSelection
: input.modelSelection;
return {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit 1753bc6. Configure here.}