Merge branch 'router-for-me:main' into main

internal/runtime/executor/usage_helpers.go

@@ -482,12 +482,16 @@ func StripUsageMetadataFromJSON(rawJSON []byte) ([]byte, bool) {
 	cleaned := jsonBytes
 	var changed bool
 
-	if gjson.GetBytes(cleaned, "usageMetadata").Exists() {
+	if usageMetadata = gjson.GetBytes(cleaned, "usageMetadata"); usageMetadata.Exists() {
+		// Rename usageMetadata to cpaUsageMetadata in the message_start event of Claude
+		cleaned, _ = sjson.SetRawBytes(cleaned, "cpaUsageMetadata", []byte(usageMetadata.Raw))
 		cleaned, _ = sjson.DeleteBytes(cleaned, "usageMetadata")
 		changed = true
 	}
 
-	if gjson.GetBytes(cleaned, "response.usageMetadata").Exists() {
+	if usageMetadata = gjson.GetBytes(cleaned, "response.usageMetadata"); usageMetadata.Exists() {
+		// Rename usageMetadata to cpaUsageMetadata in the message_start event of Claude
+		cleaned, _ = sjson.SetRawBytes(cleaned, "response.cpaUsageMetadata", []byte(usageMetadata.Raw))
 		cleaned, _ = sjson.DeleteBytes(cleaned, "response.usageMetadata")
 		changed = true
 	}

internal/translator/antigravity/claude/antigravity_claude_response.go

@@ -99,6 +99,14 @@ func ConvertAntigravityResponseToClaude(_ context.Context, _ string, originalReq
 		// This follows the Claude Code API specification for streaming message initialization
 		messageStartTemplate := `{"type": "message_start", "message": {"id": "msg_1nZdL29xx5MUA1yADyHTEsnR8uuvGzszyY", "type": "message", "role": "assistant", "content": [], "model": "claude-3-5-sonnet-20241022", "stop_reason": null, "stop_sequence": null, "usage": {"input_tokens": 0, "output_tokens": 0}}}`
 
+		// Use cpaUsageMetadata within the message_start event for Claude.
+		if promptTokenCount := gjson.GetBytes(rawJSON, "response.cpaUsageMetadata.promptTokenCount"); promptTokenCount.Exists() {
+			messageStartTemplate, _ = sjson.Set(messageStartTemplate, "message.usage.input_tokens", promptTokenCount.Int())
+		}
+		if candidatesTokenCount := gjson.GetBytes(rawJSON, "response.cpaUsageMetadata.candidatesTokenCount"); candidatesTokenCount.Exists() {
+			messageStartTemplate, _ = sjson.Set(messageStartTemplate, "message.usage.output_tokens", candidatesTokenCount.Int())
+		}
+
 		// Override default values with actual response metadata if available from the Gemini CLI response
 		if modelVersionResult := gjson.GetBytes(rawJSON, "response.modelVersion"); modelVersionResult.Exists() {
 			messageStartTemplate, _ = sjson.Set(messageStartTemplate, "message.model", modelVersionResult.String())

sdk/cliproxy/auth/conductor.go

@@ -263,7 +263,6 @@ func (m *Manager) Execute(ctx context.Context, providers []string, req cliproxye
 		return cliproxyexecutor.Response{}, &Error{Code: "provider_not_found", Message: "no provider supplied"}
 	}
 	rotated := m.rotateProviders(req.Model, normalized)
-	defer m.advanceProviderCursor(req.Model, normalized)
 
 	retryTimes, maxWait := m.retrySettings()
 	attempts := retryTimes + 1
@@ -302,7 +301,6 @@ func (m *Manager) ExecuteCount(ctx context.Context, providers []string, req clip
 		return cliproxyexecutor.Response{}, &Error{Code: "provider_not_found", Message: "no provider supplied"}
 	}
 	rotated := m.rotateProviders(req.Model, normalized)
-	defer m.advanceProviderCursor(req.Model, normalized)
 
 	retryTimes, maxWait := m.retrySettings()
 	attempts := retryTimes + 1
@@ -341,7 +339,6 @@ func (m *Manager) ExecuteStream(ctx context.Context, providers []string, req cli
 		return nil, &Error{Code: "provider_not_found", Message: "no provider supplied"}
 	}
 	rotated := m.rotateProviders(req.Model, normalized)
-	defer m.advanceProviderCursor(req.Model, normalized)
 
 	retryTimes, maxWait := m.retrySettings()
 	attempts := retryTimes + 1
@@ -640,13 +637,20 @@ func (m *Manager) normalizeProviders(providers []string) []string {
 	return result
 }
 
+// rotateProviders returns a rotated view of the providers list starting from the
+// current offset for the model, and atomically increments the offset for the next call.
+// This ensures concurrent requests get different starting providers.
 func (m *Manager) rotateProviders(model string, providers []string) []string {
 	if len(providers) == 0 {
 		return nil
 	}
-	m.mu.RLock()
+
+	// Atomic read-and-increment: get current offset and advance cursor in one lock
+	m.mu.Lock()
 	offset := m.providerOffsets[model]
-	m.mu.RUnlock()
+	m.providerOffsets[model] = (offset + 1) % len(providers)
+	m.mu.Unlock()
+
 	if len(providers) > 0 {
 		offset %= len(providers)
 	}
@@ -662,19 +666,6 @@ func (m *Manager) rotateProviders(model string, providers []string) []string {
 	return rotated
 }
 
-func (m *Manager) advanceProviderCursor(model string, providers []string) {
-	if len(providers) == 0 {
-		m.mu.Lock()
-		delete(m.providerOffsets, model)
-		m.mu.Unlock()
-		return
-	}
-	m.mu.Lock()
-	current := m.providerOffsets[model]
-	m.providerOffsets[model] = (current + 1) % len(providers)
-	m.mu.Unlock()
-}
-
 func (m *Manager) retrySettings() (int, time.Duration) {
 	if m == nil {
 		return 0, 0