feat: Abort Signal Pass-Through Providers (Group A) — Issue #616

src/api/providers/__tests__/deepseek.spec.ts

@@ -494,7 +494,7 @@ describe("DeepSeekHandler", () => {
 				expect.objectContaining({
 					thinking: { type: "enabled" },
 				}),
-				{}, // Empty path options for non-Azure URLs
+				undefined, // No signal, non-Azure URL
 			)
 			const callArgs = mockCreate.mock.calls[0][0]
 			expect(callArgs.reasoning_effort).toBeUndefined()
@@ -517,7 +517,7 @@ describe("DeepSeekHandler", () => {
 					reasoning_effort: "high",
 					max_completion_tokens: 200_000,
 				}),
-				{},
+				undefined,
 			)
 		})
 
@@ -554,7 +554,7 @@ describe("DeepSeekHandler", () => {
 					thinking: { type: "enabled" },
 					reasoning_effort: "max",
 				}),
-				{},
+				undefined,
 			)
 		})
 

src/api/providers/__tests__/lmstudio.spec.ts

@@ -133,12 +133,15 @@ describe("LmStudioHandler", () => {
 		it("should complete prompt successfully", async () => {
 			const result = await handler.completePrompt("Test prompt")
 			expect(result).toBe("Test response")
-			expect(mockCreate).toHaveBeenCalledWith({
-				model: mockOptions.lmStudioModelId,
-				messages: [{ role: "user", content: "Test prompt" }],
-				temperature: 0,
-				stream: false,
-			})
+			expect(mockCreate).toHaveBeenCalledWith(
+				{
+					model: mockOptions.lmStudioModelId,
+					messages: [{ role: "user", content: "Test prompt" }],
+					temperature: 0,
+					stream: false,
+				},
+				undefined,
+			)
 		})
 
 		it("should handle API errors", async () => {

src/api/providers/__tests__/mimo.spec.ts

@@ -376,6 +376,7 @@ describe("MimoHandler", () => {
 				expect.objectContaining({
 					extra_body: { thinking: { type: "enabled" } },
 				}),
+				undefined,
 			)
 		})
 

src/api/providers/__tests__/minimax.spec.ts

@@ -305,6 +305,7 @@ describe("MiniMaxHandler", () => {
 					messages: expect.any(Array),
 					stream: true,
 				}),
+				undefined,
 			)
 		})
 
@@ -324,6 +325,7 @@ describe("MiniMaxHandler", () => {
 				expect.objectContaining({
 					temperature: 1,
 				}),
+				undefined,
 			)
 		})
 

src/api/providers/__tests__/requesty.spec.ts

@@ -210,6 +210,7 @@ describe("RequestyHandler", () => {
 					stream_options: { include_usage: true },
 					temperature: 0,
 				}),
+				undefined,
 			)
 		})
 
@@ -243,6 +244,7 @@ describe("RequestyHandler", () => {
 					thinking: { type: "adaptive" },
 					temperature: undefined,
 				}),
+				undefined,
 			)
 		})
 
@@ -314,6 +316,7 @@ describe("RequestyHandler", () => {
 						]),
 						tool_choice: "auto",
 					}),
+					undefined,
 				)
 			})
 
@@ -412,12 +415,15 @@ describe("RequestyHandler", () => {
 
 			expect(result).toBe("test completion")
 
-			expect(mockCreate).toHaveBeenCalledWith({
-				model: mockOptions.requestyModelId,
-				max_tokens: 8192,
-				messages: [{ role: "system", content: "test prompt" }],
-				temperature: 0,
-			})
+			expect(mockCreate).toHaveBeenCalledWith(
+				{
+					model: mockOptions.requestyModelId,
+					max_tokens: 8192,
+					messages: [{ role: "system", content: "test prompt" }],
+					temperature: 0,
+				},
+				undefined,
+			)
 		})
 
 		it("omits temperature for Claude Fable 5 in completePrompt", async () => {
@@ -429,12 +435,15 @@ describe("RequestyHandler", () => {
 
 			await handler.completePrompt("test prompt")
 
-			expect(mockCreate).toHaveBeenCalledWith({
-				model: "anthropic/claude-fable-5",
-				max_tokens: 8192,
-				messages: [{ role: "system", content: "test prompt" }],
-				temperature: undefined,
-			})
+			expect(mockCreate).toHaveBeenCalledWith(
+				{
+					model: "anthropic/claude-fable-5",
+					max_tokens: 8192,
+					messages: [{ role: "system", content: "test prompt" }],
+					temperature: undefined,
+				},
+				undefined,
+			)
 		})
 
 		it("handles API errors", async () => {

src/api/providers/__tests__/unbound.spec.ts

@@ -90,6 +90,7 @@ describe("UnboundHandler", () => {
 					mode: "architect",
 				},
 			}),
+			undefined,
 		)
 	})
 })

src/api/providers/__tests__/vercel-ai-gateway.spec.ts

@@ -263,6 +263,7 @@ describe("VercelAiGatewayHandler", () => {
 				expect.objectContaining({
 					temperature: customTemp,
 				}),
+				undefined,
 			)
 		})
 
@@ -278,6 +279,7 @@ describe("VercelAiGatewayHandler", () => {
 				expect.objectContaining({
 					temperature: VERCEL_AI_GATEWAY_DEFAULT_TEMPERATURE,
 				}),
+				undefined,
 			)
 		})
 
@@ -295,6 +297,7 @@ describe("VercelAiGatewayHandler", () => {
 					temperature: undefined,
 					max_completion_tokens: 128000,
 				}),
+				undefined,
 			)
 		})
 
@@ -325,6 +328,7 @@ describe("VercelAiGatewayHandler", () => {
 				expect.objectContaining({
 					max_completion_tokens: 64000, // max tokens for sonnet 4
 				}),
+				undefined,
 			)
 		})
 
@@ -403,6 +407,7 @@ describe("VercelAiGatewayHandler", () => {
 							}),
 						]),
 					}),
+					undefined,
 				)
 			})
 
@@ -420,6 +425,7 @@ describe("VercelAiGatewayHandler", () => {
 					expect.objectContaining({
 						tool_choice: "auto",
 					}),
+					undefined,
 				)
 			})
 
@@ -437,6 +443,7 @@ describe("VercelAiGatewayHandler", () => {
 					expect.objectContaining({
 						parallel_tool_calls: true,
 					}),
+					undefined,
 				)
 			})
 
@@ -454,6 +461,7 @@ describe("VercelAiGatewayHandler", () => {
 						tools: expect.any(Array),
 						parallel_tool_calls: true,
 					}),
+					undefined,
 				)
 			})
 
@@ -553,6 +561,7 @@ describe("VercelAiGatewayHandler", () => {
 					expect.objectContaining({
 						stream_options: { include_usage: true },
 					}),
+					undefined,
 				)
 			})
 		})
@@ -591,6 +600,7 @@ describe("VercelAiGatewayHandler", () => {
 					temperature: VERCEL_AI_GATEWAY_DEFAULT_TEMPERATURE,
 					max_completion_tokens: 64000,
 				}),
+				undefined,
 			)
 		})
 
@@ -607,6 +617,7 @@ describe("VercelAiGatewayHandler", () => {
 				expect.objectContaining({
 					temperature: customTemp,
 				}),
+				undefined,
 			)
 		})
 
@@ -655,6 +666,7 @@ describe("VercelAiGatewayHandler", () => {
 				expect.objectContaining({
 					temperature: 0.9,
 				}),
+				undefined,
 			)
 		})
 	})

src/api/providers/__tests__/zai.spec.ts

@@ -502,6 +502,7 @@ describe("ZAiHandler", () => {
 					model: "glm-5.1",
 					max_tokens: 40_000,
 				}),
+				undefined,
 			)
 		})
 
@@ -542,6 +543,7 @@ describe("ZAiHandler", () => {
 					model: "glm-5.1",
 					max_tokens: 100_000,
 				}),
+				undefined,
 			)
 		})
 
@@ -572,6 +574,7 @@ describe("ZAiHandler", () => {
 					model: "glm-4.7",
 					thinking: { type: "enabled" },
 				}),
+				undefined,
 			)
 		})
 
@@ -603,6 +606,7 @@ describe("ZAiHandler", () => {
 					model: "glm-4.7",
 					thinking: { type: "disabled" },
 				}),
+				undefined,
 			)
 		})
 
@@ -634,6 +638,7 @@ describe("ZAiHandler", () => {
 					model: "glm-4.7",
 					thinking: { type: "enabled" },
 				}),
+				undefined,
 			)
 		})
 
@@ -687,6 +692,7 @@ describe("ZAiHandler", () => {
 					model: "glm-5-turbo",
 					thinking: { type: "enabled" },
 				}),
+				undefined,
 			)
 		})
 
@@ -717,6 +723,7 @@ describe("ZAiHandler", () => {
 					model: "glm-5-turbo",
 					thinking: { type: "disabled" },
 				}),
+				undefined,
 			)
 		})
 	})

src/api/providers/deepseek.ts

@@ -133,7 +133,11 @@ export class DeepSeekHandler extends OpenAiHandler {
 		try {
 			stream = await this.client.chat.completions.create(
 				requestOptions as OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming,
-				isAzureAiInference ? { path: OPENAI_AZURE_AI_INFERENCE_PATH } : {},
+				metadata?.abortSignal
+					? isAzureAiInference
+						? { path: OPENAI_AZURE_AI_INFERENCE_PATH, signal: metadata.abortSignal }
+						: { signal: metadata.abortSignal }
+					: undefined,
 			)
 		} catch (error) {
 			const { handleOpenAIError } = await import("./utils/openai-error-handler")

src/api/providers/lm-studio.ts

@@ -99,7 +99,10 @@ export class LmStudioHandler extends BaseProvider implements SingleCompletionHan
 
 			let results
 			try {
-				results = await this.client.chat.completions.create(params)
+				results = await this.client.chat.completions.create(
+					params,
+					metadata?.abortSignal ? { signal: metadata.abortSignal } : undefined,
+				)
 			} catch (error) {
 				throw handleOpenAIError(error, this.providerName)
 			}
@@ -185,7 +188,7 @@ export class LmStudioHandler extends BaseProvider implements SingleCompletionHan
 		}
 	}
 
-	async completePrompt(prompt: string): Promise<string> {
+	async completePrompt(prompt: string, metadata?: ApiHandlerCreateMessageMetadata): Promise<string> {
 		try {
 			// Create params object with optional draft model
 			const params: any = {
@@ -202,7 +205,10 @@ export class LmStudioHandler extends BaseProvider implements SingleCompletionHan
 
 			let response
 			try {
-				response = await this.client.chat.completions.create(params)
+				response = await this.client.chat.completions.create(
+					params,
+					metadata?.abortSignal ? { signal: metadata.abortSignal } : undefined,
+				)
 			} catch (error) {
 				throw handleOpenAIError(error, this.providerName)
 			}

src/api/providers/mimo.ts

@@ -99,7 +99,10 @@ export class MimoHandler extends OpenAiHandler {
 
 		let stream: AsyncIterable<OpenAI.Chat.Completions.ChatCompletionChunk>
 		try {
-			stream = (await this.client.chat.completions.create(params as any)) as any
+			stream = (await this.client.chat.completions.create(
+				params as any,
+				metadata?.abortSignal ? { signal: metadata.abortSignal } : undefined,
+			)) as any
 		} catch (error) {
 			throw handleProviderError(error, "MiMo")
 		}

src/api/providers/minimax.ts

@@ -113,7 +113,10 @@ export class MiniMaxHandler extends BaseProvider implements SingleCompletionHand
 			tool_choice: convertOpenAIToolChoice(metadata?.tool_choice),
 		}
 
-		stream = await this.client.messages.create(requestParams)
+		stream = await this.client.messages.create(
+			requestParams,
+			metadata?.abortSignal ? { signal: metadata.abortSignal } : undefined,
+		)
 
 		let inputTokens = 0
 		let outputTokens = 0
@@ -289,16 +292,19 @@ export class MiniMaxHandler extends BaseProvider implements SingleCompletionHand
 		}
 	}
 
-	async completePrompt(prompt: string) {
+	async completePrompt(prompt: string, metadata?: ApiHandlerCreateMessageMetadata) {
 		const { id: model, temperature } = this.getModel()
 
-		const message = await this.client.messages.create({
-			model,
-			max_tokens: 16_384,
-			temperature: temperature ?? 1.0,
-			messages: [{ role: "user", content: prompt }],
-			stream: false,
-		})
+		const message = await this.client.messages.create(
+			{
+				model,
+				max_tokens: 16_384,
+				temperature: temperature ?? 1.0,
+				messages: [{ role: "user", content: prompt }],
+				stream: false,
+			},
+			metadata?.abortSignal ? { signal: metadata.abortSignal } : undefined,
+		)
 
 		const content = message.content.find(({ type }) => type === "text")
 		return content?.type === "text" ? content.text : ""