跳到主要内容

toAISdkStream()

将 Mastra stream（Agent、network 或 Workflow）转换为 AI SDK 兼容的 stream。当需要手动转换 Mastra stream，以便配合 AI SDK 的 createUIMessageStream() 和 createUIMessageStreamResponse() 使用时，请使用此函数。

在 Mastra 提供的 chatRoute() 或 workflowRoute() 等路由 helper 之外构建自定义流式 endpoint 时，此函数非常有用。

toAISdkStream() 保留现有的 AI SDK v5 默认行为。如果应用使用 AI SDK v6 类型，请在 options 对象中传入 version: 'v6'。

UI stream 中的结构化输出
UI stream 中的结构化输出的直接链接

当源 Agent stream 包含最终结构化输出对象时，toAISdkStream() 会将其作为自定义 AI SDK UI data part 发送：

{
  "type": "data-structured-output",
  "data": {
    "object": {}
  }
}

object 字段包含完整的结构化输出值。它会将 Mastra 的最终结构化输出 chunk 映射到 AI SDK UI stream；不会发送部分结构化输出 chunk。

使用示例
使用示例的直接链接

Next.js App Router 示例：

app/api/chat/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { messages } = await req.json()
  const myAgent = mastra.getAgent('weatherAgent')
  const stream = await myAgent.stream(messages)

  const uiMessageStream = createUIMessageStream({
    originalMessages: messages,
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, { from: 'agent' })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}
提示

在 createUIMessageStream() 中将 messages 传给 originalMessages，可避免 UI 中出现重复的 assistant 消息。详情请参阅故障排除：assistant 消息重复。

参数
参数的直接链接

第一个参数是要转换的 Mastra stream，可以是以下类型之一：

  • MastraModelOutput — 来自 agent.stream() 的 Agent stream
  • MastraAgentNetworkStream — 来自 agent.network() 的 network stream
  • MastraWorkflowStream 或 WorkflowRunOutput — Workflow stream

第二个参数是 options 对象：

version?:

'v5' | 'v6'
= 'v5'
选择要输出的 AI SDK stream 协议。省略此项或传入 'v5' 可使用现有默认行为；当应用使用 AI SDK v6 response helper 的类型时，传入 'v6'。

from:

'agent' | 'network' | 'workflow'
= 'agent'
正在转换的 Mastra stream 类型。

lastMessageId?:

string
（仅 Agent）对话中最后一条消息的 ID。

sendStart?:

boolean
= true
（仅 Agent）是否在 stream 中发送开始事件。

sendFinish?:

boolean
= true
（仅 Agent）是否在 stream 中发送结束事件。

sendReasoning?:

boolean
= false
（仅 Agent）是否在 stream 中包含 reasoning-delta chunk。设为 true 可流式传输支持 extended thinking 的模型所生成的推理内容。

sendSources?:

boolean
= false
（仅 Agent）是否在输出中包含来源引用。

includeTextStreamParts?:

boolean
= true
（仅 Workflow）是否在输出中包含文本 stream part。

messageMetadata?:

(options: { part: UIMessageStreamPart }) => Record<string, unknown> | undefined
（仅 Agent）接收当前 stream part 并返回 metadata 的函数，返回值将附加到开始和结束 chunk。

onError?:

(error: unknown) => string
（仅 Agent）用于处理 stream 转换期间错误的函数。它接收错误，并应返回其字符串表示。

示例
示例的直接链接

转换 Workflow stream
转换 Workflow stream的直接链接

app/api/workflow/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { input } = await req.json()
  const workflow = mastra.getWorkflow('myWorkflow')
  const run = workflow.createRun()
  const stream = await run.stream({ inputData: input })

  const uiMessageStream = createUIMessageStream({
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, { from: 'workflow' })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}

转换 network stream
转换 network stream的直接链接

app/api/network/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { messages } = await req.json()
  const routingAgent = mastra.getAgent('routingAgent')
  const stream = await routingAgent.network(messages)

  const uiMessageStream = createUIMessageStream({
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, { from: 'network' })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}

转换已启用推理的 Agent stream
转换已启用推理的 Agent stream的直接链接

app/api/reasoning/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { messages } = await req.json()
  const reasoningAgent = mastra.getAgent('reasoningAgent')
  const stream = await reasoningAgent.stream(messages, {
    providerOptions: {
      openai: { reasoningEffort: 'high' },
    },
  })

  const uiMessageStream = createUIMessageStream({
    originalMessages: messages,
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, {
        from: 'agent',
        sendReasoning: true,
      })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}

为 AI SDK v6 转换 Agent stream
为 AI SDK v6 转换 Agent stream的直接链接

app/api/chat-v6/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { messages } = await req.json()
  const myAgent = mastra.getAgent('weatherAgent')
  const stream = await myAgent.stream(messages)

  const uiMessageStream = createUIMessageStream({
    originalMessages: messages,
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, {
        from: 'agent',
        version: 'v6',
      })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}

使用 messageMetadata
using-messagemetadata的直接链接

app/api/chat-with-metadata/route.ts
import { mastra } from '../../mastra'
import { createUIMessageStream, createUIMessageStreamResponse } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'

export async function POST(req: Request) {
  const { messages } = await req.json()
  const myAgent = mastra.getAgent('weatherAgent')
  const stream = await myAgent.stream(messages)

  const uiMessageStream = createUIMessageStream({
    originalMessages: messages,
    execute: async ({ writer }) => {
      for await (const part of toAISdkStream(stream, {
        from: 'agent',
        messageMetadata: ({ part }) => ({
          timestamp: Date.now(),
          partType: part.type,
        }),
      })) {
        await writer.write(part)
      }
    },
  })

  return createUIMessageStreamResponse({
    stream: uiMessageStream,
  })
}

客户端 stream 转换
客户端 stream 转换的直接链接

如果在客户端使用 Mastra client SDK（@mastra/client-js），并希望将 stream 转换为 AI SDK 格式：

client-stream-to-ai-sdk.ts
import { MastraClient } from '@mastra/client-js'
import { createUIMessageStream } from 'ai'
import { toAISdkStream } from '@mastra/ai-sdk'
import type { ChunkType, MastraModelOutput } from '@mastra/core/stream'

const client = new MastraClient({
  baseUrl: 'http://localhost:4111',
})

const agent = client.getAgent('weatherAgent')
const response = await agent.stream('What is the weather in Tokyo?')

// Convert the client SDK stream to a ReadableStream<ChunkType>
const chunkStream = new ReadableStream<ChunkType>({
  async start(controller) {
    await response.processDataStream({
      onChunk: async chunk => {
        controller.enqueue(chunk)
      },
    })
    controller.close()
  },
})

// Transform to AI SDK format
const uiMessageStream = createUIMessageStream({
  execute: async ({ writer }) => {
    for await (const part of toAISdkStream(chunkStream as unknown as MastraModelOutput, {
      from: 'agent',
    })) {
      await writer.write(part)
    }
  },
})

for await (const part of uiMessageStream) {
  console.log(part)
}
On this page