feat: enable mergeToolResultText for Roo Code Cloud provider (#10301)

src/api/providers/roo.ts

@@ -100,7 +100,13 @@ export class RooHandler extends BaseOpenAiCompatibleProvider<string> {
 			model,
 			max_tokens,
 			temperature,
-			messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)],
+			// Enable mergeToolResultText to merge environment_details and other text content
+			// after tool_results into the last tool message. This prevents reasoning/thinking
+			// models from dropping reasoning_content when they see a user message after tool results.
+			messages: [
+				{ role: "system", content: systemPrompt },
+				...convertToOpenAiMessages(messages, { mergeToolResultText: true }),
+			],
 			stream: true,
 			stream_options: { include_usage: true },
 			...(reasoning && { reasoning }),

src/api/transform/__tests__/openai-format.spec.ts

@@ -224,4 +224,181 @@ describe("convertToOpenAiMessages", () => {
 		const assistantMessage = openAiMessages[0] as OpenAI.Chat.ChatCompletionAssistantMessageParam
 		expect(assistantMessage.tool_calls![0].id).toBe("custom_toolu_123")
 	})
+
+	describe("mergeToolResultText option", () => {
+		it("should merge text content into last tool message when mergeToolResultText is true", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "tool_result",
+							tool_use_id: "tool-123",
+							content: "Tool result content",
+						},
+						{
+							type: "text",
+							text: "<environment_details>\nSome context\n</environment_details>",
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, { mergeToolResultText: true })
+
+			// Should produce only one tool message with merged content
+			expect(openAiMessages).toHaveLength(1)
+			const toolMessage = openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam
+			expect(toolMessage.role).toBe("tool")
+			expect(toolMessage.tool_call_id).toBe("tool-123")
+			expect(toolMessage.content).toBe(
+				"Tool result content\n\n<environment_details>\nSome context\n</environment_details>",
+			)
+		})
+
+		it("should merge text into last tool message when multiple tool results exist", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "tool_result",
+							tool_use_id: "call_1",
+							content: "First result",
+						},
+						{
+							type: "tool_result",
+							tool_use_id: "call_2",
+							content: "Second result",
+						},
+						{
+							type: "text",
+							text: "<environment_details>Context</environment_details>",
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, { mergeToolResultText: true })
+
+			// Should produce two tool messages, with text merged into the last one
+			expect(openAiMessages).toHaveLength(2)
+			expect((openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam).content).toBe("First result")
+			expect((openAiMessages[1] as OpenAI.Chat.ChatCompletionToolMessageParam).content).toBe(
+				"Second result\n\n<environment_details>Context</environment_details>",
+			)
+		})
+
+		it("should NOT merge text when images are present (fall back to user message)", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "tool_result",
+							tool_use_id: "tool-123",
+							content: "Tool result content",
+						},
+						{
+							type: "image",
+							source: {
+								type: "base64",
+								media_type: "image/png",
+								data: "base64data",
+							},
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, { mergeToolResultText: true })
+
+			// Should produce a tool message AND a user message (because image is present)
+			expect(openAiMessages).toHaveLength(2)
+			expect((openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam).role).toBe("tool")
+			expect(openAiMessages[1].role).toBe("user")
+		})
+
+		it("should create separate user message when mergeToolResultText is false", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "tool_result",
+							tool_use_id: "tool-123",
+							content: "Tool result content",
+						},
+						{
+							type: "text",
+							text: "<environment_details>\nSome context\n</environment_details>",
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, { mergeToolResultText: false })
+
+			// Should produce a tool message AND a separate user message (default behavior)
+			expect(openAiMessages).toHaveLength(2)
+			expect((openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam).role).toBe("tool")
+			expect((openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam).content).toBe(
+				"Tool result content",
+			)
+			expect(openAiMessages[1].role).toBe("user")
+		})
+
+		it("should work with normalizeToolCallId when mergeToolResultText is true", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "tool_result",
+							tool_use_id: "call_5019f900a247472bacde0b82",
+							content: "Tool result content",
+						},
+						{
+							type: "text",
+							text: "<environment_details>Context</environment_details>",
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, {
+				mergeToolResultText: true,
+				normalizeToolCallId: normalizeMistralToolCallId,
+			})
+
+			// Should merge AND normalize the ID
+			expect(openAiMessages).toHaveLength(1)
+			const toolMessage = openAiMessages[0] as OpenAI.Chat.ChatCompletionToolMessageParam
+			expect(toolMessage.role).toBe("tool")
+			expect(toolMessage.tool_call_id).toBe(normalizeMistralToolCallId("call_5019f900a247472bacde0b82"))
+			expect(toolMessage.content).toBe(
+				"Tool result content\n\n<environment_details>Context</environment_details>",
+			)
+		})
+
+		it("should handle user messages with only text content (no tool results)", () => {
+			const anthropicMessages: Anthropic.Messages.MessageParam[] = [
+				{
+					role: "user",
+					content: [
+						{
+							type: "text",
+							text: "Hello, how are you?",
+						},
+					],
+				},
+			]
+
+			const openAiMessages = convertToOpenAiMessages(anthropicMessages, { mergeToolResultText: true })
+
+			// Should produce a normal user message
+			expect(openAiMessages).toHaveLength(1)
+			expect(openAiMessages[0].role).toBe("user")
+		})
+	})
 })

src/api/transform/openai-format.ts

@@ -11,6 +11,14 @@ export interface ConvertToOpenAiMessagesOptions {
 	 * This allows callers to declare provider-specific ID format requirements.
 	 */
 	normalizeToolCallId?: (id: string) => string
+	/**
+	 * If true, merge text content after tool_results into the last tool message
+	 * instead of creating a separate user message. This is critical for providers
+	 * with reasoning/thinking models (like DeepSeek-reasoner, GLM-4.7, etc.) where
+	 * a user message after tool results causes the model to drop all previous
+	 * reasoning_content. Default is false for backward compatibility.
+	 */
+	mergeToolResultText?: boolean
 }
 
 export function convertToOpenAiMessages(
@@ -95,18 +103,40 @@ export function convertToOpenAiMessages(
 
 				// Process non-tool messages
 				if (nonToolMessages.length > 0) {
-					openAiMessages.push({
-						role: "user",
-						content: nonToolMessages.map((part) => {
-							if (part.type === "image") {
-								return {
-									type: "image_url",
-									image_url: { url: `data:${part.source.media_type};base64,${part.source.data}` },
+					// Check if we should merge text into the last tool message
+					// This is critical for reasoning/thinking models where a user message
+					// after tool results causes the model to drop all previous reasoning_content
+					const hasOnlyTextContent = nonToolMessages.every((part) => part.type === "text")
+					const hasToolMessages = toolMessages.length > 0
+					const shouldMergeIntoToolMessage =
+						options?.mergeToolResultText && hasToolMessages && hasOnlyTextContent
+
+					if (shouldMergeIntoToolMessage) {
+						// Merge text content into the last tool message
+						const lastToolMessage = openAiMessages[
+							openAiMessages.length - 1
+						] as OpenAI.Chat.ChatCompletionToolMessageParam
+						if (lastToolMessage?.role === "tool") {
+							const additionalText = nonToolMessages
+								.map((part) => (part as Anthropic.TextBlockParam).text)
+								.join("\n")
+							lastToolMessage.content = `${lastToolMessage.content}\n\n${additionalText}`
+						}
+					} else {
+						// Standard behavior: add user message with text/image content
+						openAiMessages.push({
+							role: "user",
+							content: nonToolMessages.map((part) => {
+								if (part.type === "image") {
+									return {
+										type: "image_url",
+										image_url: { url: `data:${part.source.media_type};base64,${part.source.data}` },
+									}
 								}
-							}
-							return { type: "text", text: part.text }
-						}),
-					})
+								return { type: "text", text: part.text }
+							}),
+						})
+					}
 				}
 			} else if (anthropicMessage.role === "assistant") {
 				const { nonToolMessages, toolMessages } = anthropicMessage.content.reduce<{