@mastra/mcp-docs-server 1.0.0-beta.9 → 1.0.1

package/.docs/raw/workflows/control-flow.mdx

@@ -1,6 +1,8 @@
 ---
 title: "Control Flow | Workflows"
 description: "Control flow in Mastra workflows allows you to manage branching, merging, and conditions to construct workflows that meet your logic requirements."
+packages:
+  - "@mastra/core"
 ---
 
 # Control Flow
@@ -22,9 +24,8 @@ Use `.then()` to run steps in order, allowing each step to access the result of
 
 ![Chaining steps with .then()](/img/workflows/workflows-control-flow-then.jpg)
 
-```typescript {30-31} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {27-28} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({
-  //...
   inputSchema: z.object({
     message: z.string()
   }),
@@ -34,7 +35,6 @@ const step1 = createStep({
 });
 
 const step2 = createStep({
-  // ...
   inputSchema: z.object({
     formatted: z.string()
   }),
@@ -44,7 +44,6 @@ const step2 = createStep({
 });
 
 export const testWorkflow = createWorkflow({
-  // ...
   inputSchema: z.object({
     message: z.string()
   }),
@@ -59,19 +58,17 @@ export const testWorkflow = createWorkflow({
 
 ## Simultaneous steps with `.parallel()`
 
-Use `.parallel()` to run steps at the same time. Each step's `id` is used when defining a following step's `inputSchema` and becomes the key on the `inputData` object used to access the previous step's values. The outputs of parallel steps can then be referenced or combined by a following step.
+Use `.parallel()` to run steps at the same time. All parallel steps must complete before the workflow continues to the next step. Each step's `id` is used when defining a following step's `inputSchema` and becomes the key on the `inputData` object used to access the previous step's values. The outputs of parallel steps can then be referenced or combined by a following step.
 
 ![Concurrent steps with .parallel()](/img/workflows/workflows-control-flow-parallel.jpg)
 
-```typescript {42} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {39} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({
   id: "step-1",
-  // ...
 });
 
 const step2 = createStep({
   id: "step-2",
-  // ...
 });
 
 const step3 = createStep({
@@ -97,7 +94,6 @@ const step3 = createStep({
 });
 
 export const testWorkflow = createWorkflow({
-  // ...
   inputSchema: z.object({
     message: z.string()
   }),
@@ -116,7 +112,7 @@ export const testWorkflow = createWorkflow({
 
 When steps run in parallel, the output is an object where each key is the step's `id` and the value is that step's output. This allows you to access each parallel step's result independently.
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({
   id: "format-step",
   inputSchema: z.object({ message: z.string() }),
@@ -176,17 +172,22 @@ export const testWorkflow = createWorkflow({
 - The next step receives an object containing all parallel step outputs
 - You must define the `inputSchema` of the following step to match this structure
 
+:::info
+
+Visit [Choosing the right pattern](#choosing-the-right-pattern) to understand when to use `.parallel()` vs `.foreach()`.
+
+:::
+
 ## Conditional logic with `.branch()`
 
 Use `.branch()` to choose which step to run based on a condition. All steps in a branch need the same `inputSchema` and `outputSchema` because branching requires consistent schemas so workflows can follow different paths.
 
 ![Conditional branching with .branch()](/img/workflows/workflows-control-flow-branch.jpg)
 
-```typescript {33-36} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {30-33} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...})
 
 const stepA = createStep({
-  // ...
   inputSchema: z.object({
     value: z.number()
   }),
@@ -196,7 +197,6 @@ const stepA = createStep({
 });
 
 const stepB = createStep({
-  // ...
   inputSchema: z.object({
     value: z.number()
   }),
@@ -206,7 +206,6 @@ const stepB = createStep({
 });
 
 export const testWorkflow = createWorkflow({
-  // ...
   inputSchema: z.object({
     value: z.number()
   }),
@@ -226,7 +225,7 @@ export const testWorkflow = createWorkflow({
 
 When using conditional branching, only one branch executes based on which condition evaluates to `true` first. The output structure is similar to `.parallel()`, where the result is keyed by the executed step's `id`.
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({
   id: "initial-step",
   inputSchema: z.object({ value: z.number() }),
@@ -307,7 +306,7 @@ When using `.then()`, `.parallel()`, or `.branch()`, it is sometimes necessary t
 
 ![Mapping with .map()](/img/workflows/workflows-data-mapping-map.jpg)
 
-```typescript {9} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {9} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...});
 const step2 = createStep({...});
 
@@ -335,7 +334,7 @@ The `.map()` method provides additional helper functions for more complex mappin
 
 When working with `.parallel()` or `.branch()` outputs, you can use `.map()` to transform the data structure before passing it to the next step. This is especially useful when you need to flatten or restructure the output.
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 export const testWorkflow = createWorkflow({...})
   .parallel([step1, step2])
   .map(async ({ inputData }) => {
@@ -350,7 +349,7 @@ export const testWorkflow = createWorkflow({...})
 
 You can also use the helper functions provided by `.map()`:
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 export const testWorkflow = createWorkflow({...})
   .branch([
     [condition1, stepA],
@@ -378,11 +377,10 @@ Use `.dountil()` to run a step repeatedly until a condition becomes true.
 
 ![Repeating with .dountil()](/img/workflows/workflows-control-flow-dountil.jpg)
 
-```typescript {17} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {14} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...});
 
 const step2 = createStep({
-  // ...
   execute: async ({ inputData }) => {
     const { number } = inputData;
     return {
@@ -391,9 +389,7 @@ const step2 = createStep({
   }
 });
 
-export const testWorkflow = createWorkflow({
-  // ...
-})
+export const testWorkflow = createWorkflow({})
   .then(step1)
   .dountil(step2, async ({ inputData: { number } }) => number > 10)
   .commit();
@@ -405,11 +401,10 @@ Use `.dowhile()` to run a step repeatedly while a condition remains true.
 
 ![Repeating with .dowhile()](/img/workflows/workflows-control-flow-dowhile.jpg)
 
-```typescript {17} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {14} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...});
 
 const step2 = createStep({
-  // ...
   execute: async ({ inputData }) => {
     const { number } = inputData;
     return {
@@ -418,9 +413,7 @@ const step2 = createStep({
   }
 });
 
-export const testWorkflow = createWorkflow({
-  // ...
-})
+export const testWorkflow = createWorkflow({})
   .then(step1)
   .dowhile(step2, async ({ inputData: { number } }) => number < 10)
   .commit();
@@ -428,13 +421,12 @@ export const testWorkflow = createWorkflow({
 
 ### Looping with `.foreach()`
 
-Use `.foreach()` to run the same step for each item in an array. The input must be of type `array` so the loop can iterate over its values, applying the step’s logic to each one.
+Use `.foreach()` to run the same step for each item in an array. The input must be of type `array` so the loop can iterate over its values, applying the step's logic to each one. See [Choosing the right pattern](#choosing-the-right-pattern) for guidance on when to use `.foreach()` vs other methods.
 
 ![Repeating with .foreach()](/img/workflows/workflows-control-flow-foreach.jpg)
 
-```typescript {17} title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript {15} title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({
-  // ...
   inputSchema: z.string(),
   outputSchema: z.string(),
   execute: async ({ inputData }) => {
@@ -445,7 +437,6 @@ const step1 = createStep({
 const step2 = createStep({...});
 
 export const testWorkflow = createWorkflow({
-  // ...
   inputSchema: z.array(z.string()),
   outputSchema: z.array(z.string())
 })
@@ -454,11 +445,37 @@ export const testWorkflow = createWorkflow({
   .commit();
 ```
 
+#### Output structure
+
+The `.foreach()` method always returns an array containing the output of each iteration. The order of outputs matches the order of inputs.
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+const addTenStep = createStep({
+  id: "add-ten",
+  inputSchema: z.object({ value: z.number() }),
+  outputSchema: z.object({ value: z.number() }),
+  execute: async ({ inputData }) => ({
+    value: inputData.value + 10
+  })
+});
+
+export const testWorkflow = createWorkflow({
+  id: "foreach-output-example",
+  inputSchema: z.array(z.object({ value: z.number() })),
+  outputSchema: z.array(z.object({ value: z.number() }))
+})
+  .foreach(addTenStep)
+  .commit();
+
+// When executed with [{ value: 1 }, { value: 22 }, { value: 333 }]
+// Output: [{ value: 11 }, { value: 32 }, { value: 343 }]
+```
+
 #### Concurrency limits
 
 Use `concurrency` to control the number of array items processed at the same time. The default is `1`, which runs steps sequentially. Increasing the value allows `.foreach()` to process multiple items simultaneously.
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...})
 
 export const testWorkflow = createWorkflow({...})
@@ -466,6 +483,324 @@ export const testWorkflow = createWorkflow({...})
   .commit();
 ```
 
+#### Aggregating results after `.foreach()`
+
+Since `.foreach()` outputs an array, you can use `.then()` or `.map()` to aggregate or transform the results. The step following `.foreach()` receives the entire array as its input.
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+const processItemStep = createStep({
+  id: "process-item",
+  inputSchema: z.object({ value: z.number() }),
+  outputSchema: z.object({ processed: z.number() }),
+  execute: async ({ inputData }) => ({
+    processed: inputData.value * 2
+  })
+});
+
+const aggregateStep = createStep({
+  id: "aggregate",
+  // Input is an array of outputs from foreach
+  inputSchema: z.array(z.object({ processed: z.number() })),
+  outputSchema: z.object({ total: z.number() }),
+  execute: async ({ inputData }) => ({
+    // Sum all processed values
+    total: inputData.reduce((sum, item) => sum + item.processed, 0)
+  })
+});
+
+export const testWorkflow = createWorkflow({
+  id: "foreach-aggregate-example",
+  inputSchema: z.array(z.object({ value: z.number() })),
+  outputSchema: z.object({ total: z.number() })
+})
+  .foreach(processItemStep)
+  .then(aggregateStep)  // Receives the full array from foreach
+  .commit();
+
+// When executed with [{ value: 1 }, { value: 2 }, { value: 3 }]
+// After foreach: [{ processed: 2 }, { processed: 4 }, { processed: 6 }]
+// After aggregate: { total: 12 }
+```
+
+You can also use `.map()` to transform the array output:
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+export const testWorkflow = createWorkflow({...})
+  .foreach(processItemStep)
+  .map(async ({ inputData }) => ({
+    // Transform the array into a different structure
+    values: inputData.map(item => item.processed),
+    count: inputData.length
+  }))
+  .then(nextStep)
+  .commit();
+```
+
+#### Chaining multiple `.foreach()` calls
+
+When you chain `.foreach()` calls, each operates on the array output of the previous step. This is useful when each item in your array needs to be transformed by multiple steps in sequence.
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+const chunkStep = createStep({
+  id: "chunk",
+  // Takes a document, returns an array of chunks
+  inputSchema: z.object({ content: z.string() }),
+  outputSchema: z.array(z.object({ chunk: z.string() })),
+  execute: async ({ inputData }) => {
+    // Split document into chunks
+    const chunks = inputData.content.match(/.{1,100}/g) || [];
+    return chunks.map(chunk => ({ chunk }));
+  }
+});
+
+const embedStep = createStep({
+  id: "embed",
+  // Takes a single chunk, returns embedding
+  inputSchema: z.object({ chunk: z.string() }),
+  outputSchema: z.object({ embedding: z.array(z.number()) }),
+  execute: async ({ inputData }) => ({
+    embedding: [/* vector embedding */]
+  })
+});
+
+// For a single document that produces multiple chunks:
+export const singleDocWorkflow = createWorkflow({
+  id: "single-doc-rag",
+  inputSchema: z.object({ content: z.string() }),
+  outputSchema: z.array(z.object({ embedding: z.array(z.number()) }))
+})
+  .then(chunkStep)      // Returns array of chunks
+  .foreach(embedStep)   // Process each chunk -> array of embeddings
+  .commit();
+```
+
+For processing multiple documents where each produces multiple chunks, you have options:
+
+**Option 1: Process all documents in a single step with batching control**
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+const downloadAndChunkStep = createStep({
+  id: "download-and-chunk",
+  inputSchema: z.array(z.string()),  // Array of URLs
+  outputSchema: z.array(z.object({ chunk: z.string(), source: z.string() })),
+  execute: async ({ inputData: urls }) => {
+    // Control batching/parallelization within the step
+    const allChunks = [];
+    for (const url of urls) {
+      const content = await fetch(url).then(r => r.text());
+      const chunks = content.match(/.{1,100}/g) || [];
+      allChunks.push(...chunks.map(chunk => ({ chunk, source: url })));
+    }
+    return allChunks;
+  }
+});
+
+export const multiDocWorkflow = createWorkflow({...})
+  .then(downloadAndChunkStep)  // Returns flat array of all chunks
+  .foreach(embedStep, { concurrency: 10 })  // Embed each chunk in parallel
+  .commit();
+```
+
+**Option 2: Use foreach for documents, aggregate chunks, then foreach for embeddings**
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+const downloadStep = createStep({
+  id: "download",
+  inputSchema: z.string(),  // Single URL
+  outputSchema: z.object({ content: z.string(), source: z.string() }),
+  execute: async ({ inputData: url }) => ({
+    content: await fetch(url).then(r => r.text()),
+    source: url
+  })
+});
+
+const chunkDocStep = createStep({
+  id: "chunk-doc",
+  inputSchema: z.object({ content: z.string(), source: z.string() }),
+  outputSchema: z.array(z.object({ chunk: z.string(), source: z.string() })),
+  execute: async ({ inputData }) => {
+    const chunks = inputData.content.match(/.{1,100}/g) || [];
+    return chunks.map(chunk => ({ chunk, source: inputData.source }));
+  }
+});
+
+export const multiDocWorkflow = createWorkflow({
+  id: "multi-doc-rag",
+  inputSchema: z.array(z.string()),  // Array of URLs
+  outputSchema: z.array(z.object({ embedding: z.array(z.number()) }))
+})
+  .foreach(downloadStep, { concurrency: 5 })  // Download docs in parallel
+  .foreach(chunkDocStep)  // Chunk each doc -> array of chunk arrays
+  .map(async ({ inputData }) => {
+    // Flatten nested arrays: [[chunks], [chunks]] -> [chunks]
+    return inputData.flat();
+  })
+  .foreach(embedStep, { concurrency: 10 })  // Embed all chunks
+  .commit();
+```
+
+**Key points about chaining `.foreach()`:**
+- Each `.foreach()` operates on the array from the previous step
+- If a step inside `.foreach()` returns an array, the output becomes an array of arrays
+- Use `.map()` with `.flat()` to flatten nested arrays when needed
+- For complex RAG pipelines, Option 1 (handling batching in a single step) often provides better control
+
+#### Nested workflows inside foreach
+
+The step after `.foreach()` only executes after all iterations complete. If you need to run multiple sequential operations per item, use a nested workflow instead of chaining multiple `.foreach()` calls. This keeps all operations for each item together and makes the data flow clearer.
+
+```typescript title="src/mastra/workflows/test-workflow.ts"
+// Define a workflow that processes a single document
+const processDocumentWorkflow = createWorkflow({
+  id: "process-document",
+  inputSchema: z.object({ url: z.string() }),
+  outputSchema: z.object({
+    embeddings: z.array(z.array(z.number())),
+    metadata: z.object({ url: z.string(), chunkCount: z.number() })
+  })
+})
+  .then(downloadStep)      // Download the document
+  .then(chunkStep)         // Split into chunks
+  .then(embedChunksStep)   // Embed all chunks for this document
+  .then(formatResultStep)  // Format the final output
+  .commit();
+
+// Use the nested workflow inside foreach
+export const batchProcessWorkflow = createWorkflow({
+  id: "batch-process-documents",
+  inputSchema: z.array(z.object({ url: z.string() })),
+  outputSchema: z.array(z.object({
+    embeddings: z.array(z.array(z.number())),
+    metadata: z.object({ url: z.string(), chunkCount: z.number() })
+  }))
+})
+  .foreach(processDocumentWorkflow, { concurrency: 3 })
+  .commit();
+
+// Each document goes through all 4 steps before the next document starts (with concurrency: 1)
+// With concurrency: 3, up to 3 documents process their full pipelines in parallel
+```
+
+**Why use nested workflows:**
+- **Better parallelism**: With `concurrency: N`, multiple items run their full pipelines simultaneously. Chained `.foreach().foreach()` processes all items through step 1, waits, then all through step 2 - nested workflows let each item progress independently
+- All steps for one item complete together before results are collected
+- Cleaner than multiple `.foreach()` calls which create nested arrays
+- Each nested workflow execution is independent with its own data flow
+- Easier to test and reuse the per-item logic separately
+
+**How it works:**
+1. The parent workflow passes each array item to an instance of the nested workflow
+2. Each nested workflow runs its full step sequence for that item
+3. With `concurrency > 1`, multiple nested workflows execute in parallel
+4. The nested workflow's final output becomes one element in the result array
+5. After all nested workflows complete, the next step in the parent receives the full array
+
+## Choosing the right pattern
+
+Use this section as a reference for selecting the appropriate control flow method.
+
+### Quick reference
+
+| Method | Purpose | Input | Output | Concurrency |
+|--------|---------|-------|--------|-------------|
+| `.then(step)` | Sequential processing | `T` | `U` | N/A (one at a time) |
+| `.parallel([a, b])` | Different operations on same input | `T` | `{ a: U, b: V }` | All run simultaneously |
+| `.foreach(step)` | Same operation on each array item | `T[]` | `U[]` | Configurable (default: 1) |
+| `.branch([...])` | Conditional path selection | `T` | `{ selectedStep: U }` | Only one branch runs |
+
+### `.parallel()` vs `.foreach()`
+
+**Use `.parallel()` when you have one input that needs different processing:**
+
+```typescript
+// Same user data processed differently in parallel
+workflow
+  .parallel([validateStep, enrichStep, scoreStep])
+  .then(combineResultsStep)
+```
+
+**Use `.foreach()` when you have many inputs that need the same processing:**
+
+```typescript
+// Multiple URLs each processed the same way
+workflow
+  .foreach(downloadStep, { concurrency: 5 })
+  .then(aggregateStep)
+```
+
+### When to use nested workflows
+
+**Inside `.foreach()`** - when each array item needs multiple sequential steps:
+
+```typescript
+// Each document goes through a full pipeline
+const processDocWorkflow = createWorkflow({...})
+  .then(downloadStep)
+  .then(parseStep)
+  .then(embedStep)
+  .commit();
+
+workflow.foreach(processDocWorkflow, { concurrency: 3 })
+```
+
+This is cleaner than chaining `.foreach().foreach()`, which creates nested arrays.
+
+**Inside `.parallel()`** - when a parallel branch needs its own multi-step pipeline:
+
+```typescript
+const pipelineA = createWorkflow({...}).then(step1).then(step2).commit();
+const pipelineB = createWorkflow({...}).then(step3).then(step4).commit();
+
+workflow.parallel([pipelineA, pipelineB])
+```
+
+### Chaining patterns
+
+| Pattern | What happens | Common use case |
+|---------|--------------|-----------------|
+| `.then().then()` | Sequential steps | Simple pipelines |
+| `.parallel().then()` | Run in parallel, then combine | Fan-out/fan-in |
+| `.foreach().then()` | Process all items, then aggregate | Map-reduce |
+| `.foreach().foreach()` | Creates array of arrays | Avoid - use nested workflow or `.map()` with `.flat()` |
+| `.foreach(workflow)` | Full pipeline per item | Multi-step processing per array item |
+
+### Synchronization: when does the next step run?
+
+Both `.parallel()` and `.foreach()` are synchronization points. The next step in the workflow only executes after all parallel branches or all array iterations have completed.
+
+```typescript
+workflow
+  .parallel([stepA, stepB, stepC])  // All 3 run simultaneously
+  .then(combineStep)                // Waits for ALL 3 to finish before running
+  .commit();
+
+workflow
+  .foreach(processStep, { concurrency: 5 })  // Up to 5 items process at once
+  .then(aggregateStep)                       // Waits for ALL items to finish before running
+  .commit();
+```
+
+This means:
+- `.parallel()` collects all branch outputs into an object, then passes it to the next step
+- `.foreach()` collects all iteration outputs into an array, then passes it to the next step
+- There is no way to "stream" results to the next step as they complete
+
+### Concurrency behavior
+
+| Method | Behavior |
+|--------|----------|
+| `.then()` | Sequential - one step at a time |
+| `.parallel()` | All branches run simultaneously (no limit option) |
+| `.foreach()` | Controlled via `{ concurrency: N }` - default is 1 (sequential) |
+| Nested workflow in `.foreach()` | Respects parent's concurrency setting |
+
+**Performance tip:** For I/O-bound operations in `.foreach()`, increase concurrency to process items in parallel:
+
+```typescript
+// Process up to 10 items simultaneously
+workflow.foreach(fetchDataStep, { concurrency: 10 })
+```
+
 ## Loop management
 
 Loop conditions can be implemented in different ways depending on how you want the loop to end. Common patterns include checking values returned in `inputData`, setting a maximum number of iterations, or aborting execution when a limit is reached.
@@ -474,7 +809,7 @@ Loop conditions can be implemented in different ways depending on how you want t
 
 Use `iterationCount` to limit how many times a loop runs. If the count exceeds your threshold, throw an error to fail the step and stop the workflow.
 
-```typescript title="src/mastra/workflows/test-workflow.ts" showLineNumbers copy
+```typescript title="src/mastra/workflows/test-workflow.ts"
 const step1 = createStep({...});
 
 export const testWorkflow = createWorkflow({...})