Release 0.2.2 (#638)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

.changeset/calm-women-repair.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-Split artifacts use case to document generator and code generator

.changeset/cuddly-cherries-trade.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-chore: improve dev experience with nodemon

.changeset/fifty-badgers-jog.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-Fix typing check issue

.changeset/five-buses-bake.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-fix chromadb dependency issue

.changeset/olive-mangos-fly.md

@@ -1,5 +0,0 @@
----
-"@llamaindex/server": patch
----
-
-feat: add dev mode UI

.changeset/silent-ladybugs-hear.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-fix: remove dead generated ai code

.changeset/tired-regions-notice.md

@@ -1,5 +0,0 @@
----
-"create-llama": patch
----
-
-Deprecate pro mode

.github/workflows/e2e.yml

@@ -79,7 +79,7 @@ jobs:
       - uses: actions/upload-artifact@v4
         if: always()
         with:
-          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}
+          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-${{ matrix.template-types }}
           path: packages/create-llama/playwright-report/
           overwrite: true
           retention-days: 30
@@ -136,6 +136,21 @@ jobs:
         run: pnpm run pack-install
         working-directory: packages/create-llama
 
+      - name: Build server
+        run: pnpm run build
+        working-directory: packages/server
+
+      - name: Pack @llamaindex/server package
+        run: |
+          pnpm pack --pack-destination "${{ runner.temp }}"
+          if [ "${{ runner.os }}" == "Windows" ]; then
+            file=$(find "${{ runner.temp }}" -name "llamaindex-server-*.tgz" | head -n 1)
+            mv "$file" "${{ runner.temp }}/llamaindex-server.tgz"
+          else
+            mv ${{ runner.temp }}/llamaindex-server-*.tgz ${{ runner.temp }}/llamaindex-server.tgz
+          fi
+        working-directory: packages/server
+
       - name: Run Playwright tests for TypeScript
         run: pnpm run e2e:typescript
         env:
@@ -144,12 +159,13 @@ jobs:
           FRAMEWORK: ${{ matrix.frameworks }}
           DATASOURCE: ${{ matrix.datasources }}
           TEMPLATE_TYPE: ${{ matrix.template-types }}
+          SERVER_PACKAGE_PATH: ${{ runner.temp }}/llamaindex-server.tgz
         working-directory: packages/create-llama
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
-          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}
+          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}-${{ matrix.template-types }}
           path: packages/create-llama/playwright-report/
           overwrite: true
           retention-days: 30

packages/create-llama/CHANGELOG.md

@@ -1,5 +1,29 @@
 # create-llama
 
+## 0.5.16
+
+### Patch Changes
+
+- 6f75d4a: fix: unsupported language in code gen workflow
+- d0618fa: Fix LlamaCloud generate script issue
+
+## 0.5.15
+
+### Patch Changes
+
+- 527075c: Enable dev mode that allows updating code directly in the UI
+
+## 0.5.14
+
+### Patch Changes
+
+- 1df8cfb: Split artifacts use case to document generator and code generator
+- 1b5a519: chore: improve dev experience with nodemon
+- b3eb0ba: Fix typing check issue
+- 556f33c: fix chromadb dependency issue
+- 2451539: fix: remove dead generated ai code
+- 7a70390: Deprecate pro mode
+
 ## 0.5.13
 
 ### Patch Changes

packages/create-llama/e2e/shared/llamaindexserver_template.spec.ts

@@ -12,21 +12,30 @@ import { createTestDir, runCreateLlama, type AppType } from "../utils";
 const templateFramework: TemplateFramework = process.env.FRAMEWORK
   ? (process.env.FRAMEWORK as TemplateFramework)
   : "fastapi";
-const dataSource: string = "--example-file";
+const dataSource: string = process.env.DATASOURCE
+  ? (process.env.DATASOURCE as string)
+  : "--example-file";
+const llamaCloudProjectName = "create-llama";
+const llamaCloudIndexName = "e2e-test";
+
 const templateUI: TemplateUI = "shadcn";
 const templatePostInstallAction: TemplatePostInstallAction = "runApp";
 const appType: AppType = "--frontend";
 const userMessage = "Write a blog post about physical standards for letters";
-const templateUseCases = ["financial_report", "agentic_rag", "deep_research"];
+const templateUseCases = [
+  "agentic_rag",
+  "financial_report",
+  "deep_research",
+  "code_generator",
+];
 
 for (const useCase of templateUseCases) {
   test.describe(`Test use case ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
     test.skip(
-      process.platform !== "linux" ||
-        process.env.DATASOURCE === "--no-files" ||
-        templateFramework === "express",
+      dataSource === "--no-files" || templateFramework === "express",
       "The llamaindexserver template currently only works with nextjs, fastapi. We also only run on Linux to speed up tests.",
     );
+    const useLlamaParse = dataSource === "--llamacloud";
     let port: number;
     let cwd: string;
     let name: string;
@@ -48,6 +57,9 @@ for (const useCase of templateUseCases) {
         templateUI,
         appType,
         useCase,
+        llamaCloudProjectName,
+        llamaCloudIndexName,
+        useLlamaParse,
       });
       name = result.projectName;
       appProcess = result.appProcess;

packages/create-llama/helpers/index.ts

@@ -18,6 +18,7 @@ import {
   ModelConfig,
   TemplateDataSource,
   TemplateFramework,
+  TemplateUseCase,
   TemplateVectorDB,
 } from "./types";
 import { installTSTemplate } from "./typescript";
@@ -60,6 +61,7 @@ async function generateContextData(
   vectorDb?: TemplateVectorDB,
   llamaCloudKey?: string,
   useLlamaParse?: boolean,
+  useCase?: TemplateUseCase,
 ) {
   if (packageManager) {
     const runGenerate = `${cyan(
@@ -96,7 +98,12 @@ async function generateContextData(
         }
       } else {
         console.log(`Running ${runGenerate} to generate the context data.`);
-        await callPackageManager(packageManager, true, ["run", "generate"]);
+        const shouldRunGenerate =
+          useCase !== "code_generator" && useCase !== "document_generator"; // Artifact use case doesn't use index.
+
+        if (shouldRunGenerate) {
+          await callPackageManager(packageManager, true, ["run", "generate"]);
+        }
         return;
       }
     }
@@ -224,6 +231,7 @@ export const installTemplate = async (
         props.vectorDb,
         props.llamaCloudKey,
         props.useLlamaParse,
+        props.useCase,
       );
     }
 

packages/create-llama/helpers/python.ts

@@ -267,7 +267,7 @@ const getAdditionalDependencies = (
     if (observability === "traceloop") {
       dependencies.push({
         name: "traceloop-sdk",
-        version: ">=0.15.11,<0.16.0",
+        version: ">=0.15.11",
       });
     }
     if (observability === "llamatrace") {
@@ -677,6 +677,7 @@ export const installPythonTemplate = async ({
     dataSources,
     tools,
     template,
+    observability,
   );
 
   await addDependencies(root, addOnDependencies);

packages/create-llama/helpers/typescript.ts

@@ -387,7 +387,7 @@ const providerDependencies: {
   [key in ModelProvider]?: Record<string, string>;
 } = {
   openai: {
-    "@llamaindex/openai": "^0.3.7",
+    "@llamaindex/openai": "~0.4.0",
   },
   gemini: {
     "@llamaindex/google": "^0.2.0",
@@ -513,7 +513,7 @@ async function updatePackageJson({
   if (backend) {
     packageJson.dependencies = {
       ...packageJson.dependencies,
-      "@llamaindex/readers": "^3.1.3",
+      "@llamaindex/readers": "~3.1.4",
     };
 
     if (vectorDb && vectorDb in vectorDbDependencies) {
@@ -543,6 +543,16 @@ async function updatePackageJson({
     };
   }
 
+  // if having custom server package tgz file, use it for testing @llamaindex/server
+  const serverPackagePath = process.env.SERVER_PACKAGE_PATH;
+  if (serverPackagePath && template === "llamaindexserver") {
+    const relativePath = path.relative(process.cwd(), serverPackagePath);
+    packageJson.dependencies = {
+      ...packageJson.dependencies,
+      "@llamaindex/server": `file:${relativePath}`,
+    };
+  }
+
   await fs.writeFile(
     packageJsonFile,
     JSON.stringify(packageJson, null, 2) + os.EOL,

packages/create-llama/package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-llama",
-  "version": "0.5.13",
+  "version": "0.5.16",
   "description": "Create LlamaIndex-powered apps with one command",
   "keywords": [
     "rag",

packages/create-llama/questions/ci.ts

@@ -6,7 +6,7 @@ const defaults: Omit<QuestionArgs, "modelConfig"> = {
   framework: "nextjs",
   ui: "shadcn",
   frontend: false,
-  llamaCloudKey: "",
+  llamaCloudKey: undefined,
   useLlamaParse: false,
   communityProjectConfig: undefined,
   llamapack: "",

packages/create-llama/templates/components/reflex/contract_review/pyproject.toml

@@ -12,7 +12,7 @@ dependencies = [
     "llama-index>=0.12.1",
     "llama-parse>=0.6.21,<0.7.0",
     "cachetools>=5.3.3",
-    "reflex>=0.6.2.post1",
+    "reflex==0.7.10",
 ]
 
 [project.scripts]

packages/create-llama/templates/components/reflex/extractor/pyproject.toml

@@ -13,7 +13,7 @@ dependencies = [
     "llama-index>=0.12.1",
     "llama-parse>=0.6.21,<0.7.0",
     "cachetools>=5.3.3",
-    "reflex>=0.6.2.post1",
+    "reflex==0.7.10",
 ]
 
 [project.scripts]

packages/create-llama/templates/components/use-cases/typescript/code_generator/src/app/workflow.ts

@@ -1,5 +1,5 @@
 import { extractLastArtifact } from "@llamaindex/server";
-import { ChatMemoryBuffer, LLM, MessageContent, Settings } from "llamaindex";
+import { ChatMemoryBuffer, MessageContent, Settings } from "llamaindex";
 
 import {
   agentStreamEvent,
@@ -12,12 +12,6 @@ import {
 
 import { z } from "zod";
 
-export const workflowFactory = async (reqBody: any) => {
-  const workflow = createCodeArtifactWorkflow(reqBody);
-
-  return workflow;
-};
-
 export const RequirementSchema = z.object({
   next_step: z.enum(["answering", "coding"]),
   language: z.string().nullable().optional(),
@@ -71,34 +65,33 @@ const artifactEvent = workflowEvent<{
   };
 }>();
 
-export function createCodeArtifactWorkflow(reqBody: any, llm?: LLM) {
-  if (!llm) {
-    llm = Settings.llm;
-  }
+export function workflowFactory(reqBody: any) {
+  const llm = Settings.llm;
+
   const { withState, getContext } = createStatefulMiddleware(() => {
     return {
-      memory: new ChatMemoryBuffer({
-        llm,
-        chatHistory: reqBody.chatHistory,
-      }),
+      memory: new ChatMemoryBuffer({ llm }),
       lastArtifact: extractLastArtifact(reqBody),
     };
   });
   const workflow = withState(createWorkflow());
 
-  workflow.handle([startAgentEvent], async ({ data: { userInput } }) => {
+  workflow.handle([startAgentEvent], async ({ data }) => {
+    const { userInput, chatHistory = [] } = data;
     // Prepare chat history
     const { state } = getContext();
     // Put user input to the memory
     if (!userInput) {
       throw new Error("Missing user input to start the workflow");
     }
-    state.memory.put({
-      role: "user",
-      content: userInput,
-    });
+    state.memory.set(chatHistory);
+    state.memory.put({ role: "user", content: userInput });
+
     return planEvent.with({
       userInput: userInput,
+      context: state.lastArtifact
+        ? JSON.stringify(state.lastArtifact)
+        : undefined,
     });
   });
 

packages/create-llama/templates/components/use-cases/typescript/document_generator/src/app/workflow.ts

@@ -1,5 +1,5 @@
 import { extractLastArtifact } from "@llamaindex/server";
-import { ChatMemoryBuffer, LLM, MessageContent, Settings } from "llamaindex";
+import { ChatMemoryBuffer, MessageContent, Settings } from "llamaindex";
 
 import {
   agentStreamEvent,
@@ -12,12 +12,6 @@ import {
 
 import { z } from "zod";
 
-export const workflowFactory = async (reqBody: any) => {
-  const workflow = createDocumentArtifactWorkflow(reqBody);
-
-  return workflow;
-};
-
 export const DocumentRequirementSchema = z.object({
   type: z.enum(["markdown", "html"]),
   title: z.string(),
@@ -74,32 +68,28 @@ const artifactEvent = workflowEvent<{
   };
 }>();
 
-export function createDocumentArtifactWorkflow(reqBody: any, llm?: LLM) {
-  if (!llm) {
-    llm = Settings.llm;
-  }
+export function workflowFactory(reqBody: any) {
+  const llm = Settings.llm;
+
   const { withState, getContext } = createStatefulMiddleware(() => {
     return {
-      memory: new ChatMemoryBuffer({
-        llm,
-        chatHistory: reqBody.chatHistory,
-      }),
+      memory: new ChatMemoryBuffer({ llm }),
       lastArtifact: extractLastArtifact(reqBody),
     };
   });
   const workflow = withState(createWorkflow());
 
-  workflow.handle([startAgentEvent], async ({ data: { userInput } }) => {
+  workflow.handle([startAgentEvent], async ({ data }) => {
+    const { userInput, chatHistory = [] } = data;
     // Prepare chat history
     const { state } = getContext();
     // Put user input to the memory
     if (!userInput) {
       throw new Error("Missing user input to start the workflow");
     }
-    state.memory.put({
-      role: "user",
-      content: userInput,
-    });
+    state.memory.set(chatHistory);
+    state.memory.put({ role: "user", content: userInput });
+
     return planEvent.with({
       userInput,
       context: state.lastArtifact

packages/create-llama/templates/components/vectordbs/llamaindexserver/llamacloud/python/generate.py

@@ -12,11 +12,12 @@ from llama_index.server.services.llamacloud.generate import (
     load_to_llamacloud,
 )
 
+
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger()
 
 
-def generate_datasource():
+def generate_index():
     init_settings()
     logger.info("Generate index for the provided data")
 
@@ -27,5 +28,26 @@ def generate_datasource():
     load_to_llamacloud(index, logger=logger)
 
 
-if __name__ == "__main__":
-    generate_datasource()
+def generate_ui_for_workflow():
+    """
+    Generate UI for UIEventData event in app/workflow.py
+    """
+    import asyncio
+    from llama_index.llms.openai import OpenAI
+    from main import COMPONENT_DIR
+
+    # To generate UI components for additional event types,
+    # import the corresponding data model (e.g., MyCustomEventData)
+    # and run the generate_ui_for_workflow function with the imported model.
+    # Make sure the output filename of the generated UI component matches the event type (here `ui_event`)
+    try:
+        from app.workflow import UIEventData  # type: ignore
+    except ImportError:
+        raise ImportError("Couldn't generate UI component for the current workflow.")
+    from llama_index.server.gen_ui import generate_event_component
+
+    # works also well with Claude 3.7 Sonnet or Gemini Pro 2.5
+    llm = OpenAI(model="gpt-4.1")
+    code = asyncio.run(generate_event_component(event_cls=UIEventData, llm=llm))
+    with open(f"{COMPONENT_DIR}/ui_event.jsx", "w") as f:
+        f.write(code)

packages/create-llama/templates/components/vectordbs/llamaindexserver/llamacloud/typescript/generate.ts

@@ -1,7 +1,8 @@
+import { generateEventComponent } from "@llamaindex/server";
 import * as dotenv from "dotenv";
 import "dotenv/config";
 import * as fs from "fs/promises";
-import { LLamaCloudFileService } from "llamaindex";
+import { LLamaCloudFileService, OpenAI } from "llamaindex";
 import * as path from "path";
 import { getIndex } from "./app/data";
 import { initSettings } from "./app/settings";
@@ -88,7 +89,7 @@ async function loadAndIndex() {
   console.log(`Successfully uploaded documents to LlamaCloud!`);
 }
 
-(async () => {
+async function generateDatasource() {
   try {
     checkRequiredEnvVars();
     initSettings();
@@ -97,4 +98,39 @@ async function loadAndIndex() {
   } catch (error) {
     console.error("Error generating storage.", error);
   }
+}
+
+async function generateUi() {
+  // Also works well with Claude 3.5 Sonnet and Google Gemini 2.5 Pro
+  const llm = new OpenAI({ model: "gpt-4.1" });
+
+  const workflowModule = await import("./app/workflow");
+  const UIEventSchema = (workflowModule as any).UIEventSchema;
+  if (!UIEventSchema) {
+    throw new Error(
+      "To generate the UI, you must define a UIEventSchema in your workflow.",
+    );
+  }
+
+  const generatedCode = await generateEventComponent(UIEventSchema, llm);
+  // Write the generated code to components/ui_event.ts
+  await fs.writeFile("components/ui_event.jsx", generatedCode);
+}
+
+(async () => {
+  const args = process.argv.slice(2);
+  const command = args[0];
+
+  initSettings();
+
+  if (command === "datasource") {
+    await generateDatasource();
+  } else if (command === "ui") {
+    await generateUi();
+  } else {
+    console.error(
+      'Invalid command. Please use "datasource" or "ui". Running "datasource" by default.',
+    );
+    await generateDatasource(); // Default behavior or could throw an error
+  }
 })();

packages/create-llama/templates/types/llamaindexserver/fastapi/app/settings.py

@@ -1,8 +1,12 @@
+import os
+
 from llama_index.core import Settings
 from llama_index.embeddings.openai import OpenAIEmbedding
 from llama_index.llms.openai import OpenAI
 
 
 def init_settings():
+    if os.getenv("OPENAI_API_KEY") is None:
+        raise RuntimeError("OPENAI_API_KEY is missing in environment variables")
     Settings.llm = OpenAI(model="gpt-4o-mini")
     Settings.embed_model = OpenAIEmbedding(model="text-embedding-3-small")

packages/create-llama/templates/types/llamaindexserver/fastapi/main.py

@@ -17,8 +17,10 @@ def create_app():
         ui_config=UIConfig(
             component_dir=COMPONENT_DIR,
             app_title="Chat App",
+            dev_mode=True,  # Please disable this in production
         ),
         logger=logger,
+        env="dev",
     )
     # You can also add custom FastAPI routes to app
     app.add_api_route("/api/health", lambda: {"message": "OK"}, status_code=200)

packages/create-llama/templates/types/llamaindexserver/fastapi/pyproject.toml

@@ -12,7 +12,7 @@ dependencies = [
     "pydantic<2.10",
     "aiostream>=0.5.2,<0.6.0",
     "llama-index-core>=0.12.28,<0.13.0",
-    "llama-index-server>=0.1.15,<0.2.0",
+    "llama-index-server>=0.1.16,<0.2.0",
 ]
 
 [project.optional-dependencies]

packages/create-llama/templates/types/llamaindexserver/nextjs/package.json

@@ -9,11 +9,11 @@
     "start": "tsx src/index.ts"
   },
   "dependencies": {
-    "@llamaindex/openai": "^0.3.7",
-    "@llamaindex/server": "^0.2.0",
-    "@llamaindex/workflow": "^1.1.2",
-    "@llamaindex/tools": "^0.0.10",
-    "llamaindex": "^0.10.6",
+    "@llamaindex/openai": "~0.4.0",
+    "@llamaindex/server": "~0.2.1",
+    "@llamaindex/workflow": "~1.1.3",
+    "@llamaindex/tools": "~0.0.11",
+    "llamaindex": "~0.11.0",
     "dotenv": "^16.4.7",
     "zod": "^3.23.8"
   },

packages/create-llama/templates/types/llamaindexserver/nextjs/src/generate.ts

@@ -1,14 +1,10 @@
 import "dotenv/config";
 import { SimpleDirectoryReader } from "@llamaindex/readers/directory";
-import {
-  OpenAI,
-  storageContextFromDefaults,
-  VectorStoreIndex,
-} from "llamaindex";
+import { storageContextFromDefaults, VectorStoreIndex } from "llamaindex";
 import { initSettings } from "./app/settings";
 import fs from "fs";
 import { generateEventComponent } from "@llamaindex/server";
-import { UIEventSchema } from "./app/workflow";
+import { OpenAI } from "@llamaindex/openai";
 
 async function generateDatasource() {
   console.log(`Generating storage context...`);
@@ -30,6 +26,14 @@ async function generateUi() {
   // Also works well with Claude 3.5 Sonnet and Google Gemini 2.5 Pro
   const llm = new OpenAI({ model: "gpt-4.1" });
 
+  const workflowModule = await import("./app/workflow");
+  const UIEventSchema = (workflowModule as any).UIEventSchema;
+  if (!UIEventSchema) {
+    throw new Error(
+      "To generate the UI, you must define a UIEventSchema in your workflow.",
+    );
+  }
+
   // You can also generate for other workflow events
   const generatedCode = await generateEventComponent(UIEventSchema, llm);
   // Write the generated code to components/ui_event.ts

packages/create-llama/templates/types/llamaindexserver/nextjs/src/index.ts

@@ -10,5 +10,6 @@ new LlamaIndexServer({
   uiConfig: {
     appTitle: "LlamaIndex App",
     componentsDir: "components",
+    devMode: true,
   },
 }).start();

packages/server/.gitignore

@@ -1 +1,5 @@
-server/
+# server contains Nextjs frontend code (not compiled)
+server/
+
+# temp is the copy of next folder but without API folder, used to build frontend static files
+temp/

packages/server/CHANGELOG.md

@@ -1,5 +1,18 @@
 # @llamaindex/server
 
+## 0.2.2
+
+### Patch Changes
+
+- 25fba43: refactor: migrate to Nextjs Route Handler
+- 6f75d4a: fix: unsupported language in code gen workflow
+
+## 0.2.1
+
+### Patch Changes
+
+- f072308: feat: add dev mode UI
+
 ## 0.2.0
 
 ### Minor Changes

packages/server/README.md

@@ -4,10 +4,10 @@ LlamaIndexServer is a Next.js-based application that allows you to quickly launc
 
 ## Features
 
-- Serving a workflow as a chatbot
+- Add a sophisticated chatbot UI to your LlamaIndex workflow
+- Edit code and document artifacts in an OpenAI Canvas-style UI
+- Extendable UI components for events and headers
 - Built on Next.js for high performance and easy API development
-- Optional built-in chat UI with extendable UI components
-- Prebuilt development code
 
 ## Installation
 
@@ -21,9 +21,11 @@ Create an `index.ts` file and add the following code:
 
 ```ts
 import { LlamaIndexServer } from "@llamaindex/server";
+import { openai } from "@llamaindex/openai";
+import { agent } from "@llamaindex/workflow";
 import { wiki } from "@llamaindex/tools"; // or any other tool
 
-const createWorkflow = () => agent({ tools: [wiki()] });
+const createWorkflow = () => agent({ tools: [wiki()], llm: openai("gpt-4o") });
 
 new LlamaIndexServer({
   workflow: createWorkflow,
@@ -34,6 +36,8 @@ new LlamaIndexServer({
 }).start();
 ```
 
+The `createWorkflow` function is a factory function that creates an [Agent Workflow](https://ts.llamaindex.ai/docs/llamaindex/modules/agents/agent_workflow) with a tool that retrieves information from Wikipedia in this case. For more details, read about the [Workflow factory contract](#workflow-factory-contract).
+
 ## Running the Server
 
 In the same directory as `index.ts`, run the following command to start the server:
@@ -54,16 +58,75 @@ curl -X POST "http://localhost:3000/api/chat" -H "Content-Type: application/json
 
 The `LlamaIndexServer` accepts the following configuration options:
 
-- `workflow`: A callable function that creates a workflow instance for each request
+- `workflow`: A callable function that creates a workflow instance for each request. See [Workflow factory contract](#workflow-factory-contract) for more details.
 - `uiConfig`: An object to configure the chat UI containing the following properties:
   - `appTitle`: The title of the application (default: `"LlamaIndex App"`)
   - `starterQuestions`: List of starter questions for the chat UI (default: `[]`)
   - `componentsDir`: The directory for custom UI components rendering events emitted by the workflow. The default is undefined, which does not render custom UI components.
   - `llamaCloudIndexSelector`: Whether to show the LlamaCloud index selector in the chat UI (requires `LLAMA_CLOUD_API_KEY` to be set in the environment variables) (default: `false`)
+  - `dev_mode`: When enabled, you can update workflow code in the UI and see the changes immediately. It's currently in beta and only supports updating workflow code at `app/src/workflow.ts`. Please start server in dev mode (`npm run dev`) to use see this reload feature enabled.
 
 LlamaIndexServer accepts all the configuration options from Nextjs Custom Server such as `port`, `hostname`, `dev`, etc.
 See all Nextjs Custom Server options [here](https://nextjs.org/docs/app/building-your-application/configuring/custom-server).
 
+## Workflow factory contract
+
+The `workflow` provided will be called for each chat request to initialize a new workflow instance. The contract of the generated workflow must be the same as for the [Agent Workflow](https://ts.llamaindex.ai/docs/llamaindex/modules/agents/agent_workflow).
+
+This means that the workflow must handle a `startAgentEvent` event, which is the entry point of the workflow and contains the following information in it's `data` property:
+
+```typescript
+{
+  userInput: MessageContent;
+  chatHistory?: ChatMessage[] | undefined;
+};
+```
+
+The `userInput` is the latest user message and the `chatHistory` is the list of messages exchanged between the user and the workflow so far.
+
+Furthermore, the workflow must stop with a `stopAgentEvent` event to mark the end of the workflow. In between, the workflow can emit [UI events](##AI-generated-UI-Components) to render custom UI components and [Artifact events](##Sending-Artifacts-to-the-UI) to send structured data like generated documents or code snippets to the UI.
+
+```ts
+import {
+  createStatefulMiddleware,
+  createWorkflow,
+  startAgentEvent,
+} from "@llamaindex/workflow";
+import { ChatMemoryBuffer, type ChatMessage, Settings } from "llamaindex";
+import { openai } from "@llamaindex/openai";
+import { wiki } from "@llamaindex/tools";
+
+Settings.llm = openai("gpt-4o");
+
+export const workflowFactory = async () => {
+  const workflow = createWorkflow();
+
+  workflow.handle([startAgentEvent], async ({ data }) => {
+    const { state, sendEvent } = getContext();
+    const messages = data.chatHistory;
+
+    const toolCallResponse = await chatWithTools(
+      Settings.llm,
+      [wiki()],
+      messages,
+    );
+
+    // using result from tool call and use `sendEvent` to emit the next event...
+  });
+
+  // define more workflow handling logic here...
+
+  // Finally stop with a `stopAgentEvent` event to mark the end of the workflow.
+  // return stopAgentEvent.with({
+  //   result: "This is the end!",
+  // });
+
+  return workflow;
+};
+```
+
+To generate sophisticated examples of workflows, you best use the [create-llama](https://github.com/run-llama/create-llama) project.
+
 ## AI-generated UI Components
 
 The LlamaIndex server provides support for rendering workflow events using custom UI components, allowing you to extend and customize the chat interface.
@@ -137,6 +200,66 @@ new LlamaIndexServer({
 }).start();
 ```
 
+## Sending Artifacts to the UI
+
+In addition to UI events for custom components, LlamaIndex Server supports a special `ArtifactEvent` to send structured data like generated documents or code snippets to the UI. These artifacts are displayed in a dedicated "Canvas" panel in the chat interface.
+
+### Artifact Event Structure
+
+To send an artifact, your workflow needs to emit an event with `type: "artifact"`. The `data` payload of this event should include:
+
+- `type`: A string indicating the type of artifact (e.g., `"document"`, `"code"`).
+- `created_at`: A timestamp (e.g., `Date.now()`) indicating when the artifact was created.
+- `data`: An object containing the specific details of the artifact. The structure of this object depends on the artifact `type`.
+
+### Defining and Sending an ArtifactEvent
+
+First, define your artifact event using `workflowEvent` from `@llamaindex/workflow`:
+
+```typescript
+import { workflowEvent } from "@llamaindex/workflow";
+
+// Example for a document artifact
+const artifactEvent = workflowEvent<{
+  type: "artifact"; // Must be "artifact"
+  data: {
+    type: "document"; // Custom type for your artifact (e.g., "document", "code")
+    created_at: number;
+    data: {
+      // Specific data for the document artifact type
+      title: string;
+      content: string;
+      type: "markdown" | "html"; // document format
+    };
+  };
+}>();
+```
+
+Then, within your workflow logic, use `sendEvent` (obtained from `getContext()`) to emit the event:
+
+```typescript
+// Assuming 'sendEvent' is available in your workflow handler
+// and 'documentDetails' contains the content for the artifact.
+
+sendEvent(
+  artifactEvent.with({
+    type: "artifact", // This top-level type must be "artifact"
+    data: {
+      type: "document", // This is your specific artifact type
+      created_at: Date.now(),
+      data: {
+        title: "My Generated Document",
+        content: "# Hello World
+This is a markdown document.",
+        type: "markdown",
+      },
+    },
+  }),
+);
+```
+
+This will send the artifact to the LlamaIndex Server UI, where it will be rendered in the [ChatCanvasPanel](/packages/server/next/app/components/ui/chat/canvas/panel.tsx) by a renderer depending on the artifact type. For type `document` this is using the [DocumentArtifactViewer](https://github.com/run-llama/chat-ui/blob/bacb75fc6edceacf742fba18632404a2483b5a81/packages/chat-ui/src/chat/canvas/artifacts/document.tsx#L17).
+
 ## Default Endpoints and Features
 
 ### Chat Endpoint

packages/server/examples/README.md

@@ -6,7 +6,7 @@ This directory contains examples of how to use the LlamaIndex Server.
 
 ```bash
 export OPENAI_API_KEY=your_openai_api_key
-npx tsx simple-workflow/calculator.ts
+pnpm run dev
 ```
 
 ## Open browser at http://localhost:3000

packages/server/examples/agentic-rag/index.ts

@@ -39,5 +39,5 @@ new LlamaIndexServer({
     appTitle: "LlamaIndex App",
     starterQuestions: ["What is the color of the dog?"],
   },
-  port: 4100,
+  port: 3000,
 }).start();

packages/server/examples/devmode/README.md

@@ -9,7 +9,7 @@ new LlamaIndexServer({
     appTitle: "Calculator",
     devMode: true,
   },
-  port: 6000,
+  port: 3000,
 }).start();
 ```
 
@@ -17,5 +17,5 @@ Export OpenAI API key and start the server in dev mode.
 
 ```bash
 export OPENAI_API_KEY=<your-openai-api-key>
-npx tsx watch index.ts
+npx nodemon --exec tsx index.ts --ignore src/app/workflow_*.ts
 ```

packages/server/examples/devmode/index.ts

@@ -11,5 +11,5 @@ new LlamaIndexServer({
       "What is the weather in New York?",
     ],
   },
-  port: 6005,
+  port: 3000,
 }).start();

packages/server/examples/simple-workflow/calculator.ts

@@ -20,5 +20,5 @@ new LlamaIndexServer({
     appTitle: "Calculator",
     starterQuestions: ["1 + 1", "2 + 2"],
   },
-  port: 4000,
+  port: 3000,
 }).start();

packages/server/next/app/api/chat/config/llamacloud/route.ts

@@ -0,0 +1,32 @@
+import { getEnv } from "@llamaindex/env";
+import { LLamaCloudFileService } from "llamaindex";
+import { NextRequest, NextResponse } from "next/server";
+
+export async function GET(request: NextRequest): Promise<NextResponse> {
+  if (!getEnv("LLAMA_CLOUD_API_KEY")) {
+    return NextResponse.json(
+      {
+        error: "env variable LLAMA_CLOUD_API_KEY is required to use LlamaCloud",
+      },
+      { status: 500 },
+    );
+  }
+
+  try {
+    const config = {
+      projects: await LLamaCloudFileService.getAllProjectsWithPipelines(),
+      pipeline: {
+        pipeline: getEnv("LLAMA_CLOUD_INDEX_NAME"),
+        project: getEnv("LLAMA_CLOUD_PROJECT_NAME"),
+      },
+    };
+    return NextResponse.json(config, { status: 200 });
+  } catch (error) {
+    return NextResponse.json(
+      {
+        error: "Failed to fetch LlamaCloud configuration",
+      },
+      { status: 500 },
+    );
+  }
+}

packages/server/next/app/api/chat/route.ts

@@ -0,0 +1,77 @@
+import { type AgentInputData } from "@llamaindex/workflow";
+import { type Message } from "ai";
+import { type MessageType } from "llamaindex";
+import { NextRequest, NextResponse } from "next/server";
+
+// import chat utils
+import { toDataStream } from "./utils/stream";
+import { sendSuggestedQuestionsEvent } from "./utils/suggestion";
+import { runWorkflow } from "./utils/workflow";
+
+// import workflow factory from local file
+import { workflowFactory } from "../../../../app/workflow";
+
+export async function POST(req: NextRequest) {
+  try {
+    const reqBody = await req.json();
+    const { messages } = reqBody as { messages: Message[] };
+    const chatHistory = messages.map((message) => ({
+      role: message.role as MessageType,
+      content: message.content,
+    }));
+
+    const lastMessage = messages[messages.length - 1];
+    if (lastMessage?.role !== "user") {
+      return NextResponse.json(
+        {
+          detail: "Messages cannot be empty and last message must be from user",
+        },
+        { status: 400 },
+      );
+    }
+    const workflowInput: AgentInputData = {
+      userInput: lastMessage.content,
+      chatHistory,
+    };
+
+    const abortController = new AbortController();
+    req.signal.addEventListener("abort", () =>
+      abortController.abort("Connection closed"),
+    );
+
+    const workflow = await workflowFactory(reqBody);
+    const workflowEventStream = await runWorkflow(
+      workflow,
+      workflowInput,
+      abortController.signal,
+    );
+
+    const dataStream = toDataStream(workflowEventStream, {
+      callbacks: {
+        onFinal: async (completion, dataStreamWriter) => {
+          chatHistory.push({
+            role: "assistant" as MessageType,
+            content: completion,
+          });
+          await sendSuggestedQuestionsEvent(dataStreamWriter, chatHistory);
+        },
+      },
+    });
+
+    return new Response(dataStream, {
+      status: 200,
+      headers: {
+        "Content-Type": "text/plain; charset=utf-8",
+        "X-Vercel-AI-Data-Stream": "v1",
+      },
+    });
+  } catch (error) {
+    console.error("Chat handler error:", error);
+    return NextResponse.json(
+      {
+        detail: (error as Error).message || "Internal server error",
+      },
+      { status: 500 },
+    );
+  }
+}

packages/server/next/app/api/components/route.ts

@@ -1,20 +1,19 @@
 import fs from "fs";
-import type { IncomingMessage, ServerResponse } from "http";
+import { NextRequest, NextResponse } from "next/server";
 import path from "path";
 import { promisify } from "util";
-import { sendJSONResponse } from "../utils/request";
 
-export const getComponents = async (
-  req: IncomingMessage,
-  res: ServerResponse,
-  componentsDir: string,
-) => {
+export async function GET(request: NextRequest) {
+  const params = request.nextUrl.searchParams;
+  const componentsDir = params.get("componentsDir") || "components";
+
   try {
     const exists = await promisify(fs.exists)(componentsDir);
     if (!exists) {
-      return sendJSONResponse(res, 404, {
-        error: "Components directory not found",
-      });
+      return NextResponse.json(
+        { error: "Components directory not found" },
+        { status: 404 },
+      );
     }
 
     const files = await promisify(fs.readdir)(componentsDir);
@@ -40,12 +39,15 @@ export const getComponents = async (
       }),
     );
 
-    sendJSONResponse(res, 200, components);
+    return NextResponse.json(components, { status: 200 });
   } catch (error) {
     console.error("Error reading components:", error);
-    sendJSONResponse(res, 500, { error: "Failed to read components" });
+    return NextResponse.json(
+      { error: "Failed to read components" },
+      { status: 500 },
+    );
   }
-};
+}
 
 function filterDuplicateComponents(files: string[]) {
   const compMap = new Map<string, string>();

packages/server/next/app/api/dev/files/workflow/route.ts

@@ -0,0 +1,96 @@
+import { exec } from "child_process";
+import fs from "fs";
+import { NextRequest, NextResponse } from "next/server";
+import path from "path";
+import { promisify } from "util";
+
+const DEFAULT_WORKFLOW_FILE_PATH = "src/app/workflow.ts"; // TODO: we can make it as a parameter in server later
+
+export async function GET(request: NextRequest) {
+  const filePath = DEFAULT_WORKFLOW_FILE_PATH;
+
+  const fileExists = await promisify(fs.exists)(DEFAULT_WORKFLOW_FILE_PATH);
+  if (!fileExists) {
+    return NextResponse.json(
+      {
+        detail: `Dev mode is currently in beta. It only supports updating workflow file at ${filePath}`,
+      },
+      { status: 404 },
+    );
+  }
+
+  const content = await promisify(fs.readFile)(filePath, "utf-8");
+  const last_modified = fs.statSync(filePath).mtime.getTime();
+
+  return NextResponse.json(
+    { content, file_path: filePath, last_modified },
+    { status: 200 },
+  );
+}
+
+export async function PUT(request: NextRequest) {
+  const filePath = DEFAULT_WORKFLOW_FILE_PATH;
+  const { content } = await request.json();
+
+  const fileExists = await promisify(fs.exists)(filePath);
+  if (!fileExists) {
+    return NextResponse.json(
+      {
+        detail: `Dev mode is currently in beta. It only supports updating workflow file at ${DEFAULT_WORKFLOW_FILE_PATH}`,
+      },
+      { status: 404 },
+    );
+  }
+
+  try {
+    const resolvedFilePath = path.resolve(DEFAULT_WORKFLOW_FILE_PATH);
+    const result = await validateTypeScriptFile(resolvedFilePath, content);
+
+    if (!result.isValid) {
+      return NextResponse.json(
+        {
+          detail: result.errors.join("\n"),
+        },
+        { status: 400 },
+      );
+    }
+
+    await promisify(fs.writeFile)(filePath, content);
+    return NextResponse.json({ content }, { status: 200 });
+  } catch (error) {
+    console.error("Error updating workflow file:", error);
+    return NextResponse.json(
+      { error: "Failed to update workflow file" },
+      { status: 500 },
+    );
+  }
+}
+
+// use typescript package to validate the file syntax and imports
+async function validateTypeScriptFile(filePath: string, content: string) {
+  // Update workflow file directly will cause the server restart immediately.
+  // So we create a temporary file with the same content in the same directory as the workflow file
+  // This file will be used to validate the file syntax and imports. It will be deleted after validation.
+  const tempFilePath = path.join(
+    path.dirname(filePath),
+    `workflow_${Date.now()}.ts`,
+  );
+  fs.writeFileSync(tempFilePath, content);
+
+  const errors = [];
+  try {
+    const tscCommand = `npx tsc ${tempFilePath} --noEmit --skipLibCheck true`;
+    await promisify(exec)(tscCommand);
+  } catch (error) {
+    const errorMessage = (error as { stdout: string })?.stdout;
+    errors.push(errorMessage);
+  } finally {
+    // Clean up temporary file
+    if (fs.existsSync(tempFilePath)) fs.unlinkSync(tempFilePath);
+  }
+
+  return {
+    isValid: errors.length === 0,
+    errors: errors,
+  };
+}

packages/server/next/app/api/files/[...slug]/route.ts

@@ -0,0 +1,24 @@
+import fs from "fs";
+import { NextRequest, NextResponse } from "next/server";
+import { promisify } from "util";
+
+export async function GET(
+  request: NextRequest,
+  { params }: { params: Promise<{ slug: string[] }> },
+) {
+  const filePath = (await params).slug.join("/");
+
+  if (!filePath.startsWith("output") && !filePath.startsWith("data")) {
+    return NextResponse.json({ error: "No permission" }, { status: 400 });
+  }
+
+  const decodedFilePath = decodeURIComponent(filePath);
+  const fileExists = await promisify(fs.exists)(decodedFilePath);
+
+  if (fileExists) {
+    const fileBuffer = await promisify(fs.readFile)(decodedFilePath);
+    return new NextResponse(fileBuffer);
+  } else {
+    return NextResponse.json({ error: "File not found" }, { status: 404 });
+  }
+}

packages/server/next/app/components/ui/chat/dev-mode-panel.tsx

@@ -5,7 +5,7 @@ import {
   fileExtensionToEditorLang,
 } from "@llamaindex/chat-ui/widgets";
 import { AlertCircle, Loader2 } from "lucide-react";
-import { useEffect, useState } from "react";
+import { useEffect, useMemo, useState } from "react";
 import { Button } from "../button";
 import { getConfig } from "../lib/utils";
 
@@ -144,9 +144,12 @@ function DevModePanelComp() {
     }
   }, [devModeOpen]);
 
-  const codeEditorLanguage = fileExtensionToEditorLang(
-    workflowFile?.file_path.split(".").pop() ?? "",
-  );
+  const codeEditorLanguage = useMemo(() => {
+    if (!workflowFile?.file_path) return undefined;
+    return fileExtensionToEditorLang(
+      workflowFile.file_path.split(".").pop() ?? "",
+    );
+  }, [workflowFile]);
 
   return (
     <>

packages/server/next/tsconfig.json

@@ -20,7 +20,7 @@
     "paths": {
       "@/*": ["./*"]
     },
-    "target": "ES2017"
+    "target": "ES2022"
   },
   "include": ["next-env.d.ts", "**/*.ts", "**/*.tsx", ".next/types/**/*.ts"],
   "exclude": ["node_modules"]

packages/server/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/server",
   "description": "LlamaIndex Server",
-  "version": "0.2.0",
+  "version": "0.2.2",
   "type": "module",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",
@@ -27,21 +27,25 @@
     "directory": "packages/server"
   },
   "scripts": {
-    "clean": "rm -rf ./dist ./server next/.next next/out",
+    "dev": "bunchee --watch",
+    "clean": "rm -rf ./dist ./server next/.next next/out ./temp",
     "prebuild": "pnpm clean",
     "build": "bunchee",
-    "postbuild": "pnpm copy:next-src && pnpm build:static && pnpm copy:static",
-    "copy:next-src": "cp -r ./next ./server && pnpm build:css && rm -rf ./server/postcss.config.js",
-    "build:css": "postcss server/app/globals.css -o server/app/globals.css",
-    "build:static": "cd ./next && next build",
-    "copy:static": "cp -r ./next/out ./dist/static",
-    "dev": "bunchee --watch"
+    "postbuild": "pnpm prepare:ts-server && pnpm prepare:py-static",
+    "prepare:ts-server": "pnpm copy:next-src && pnpm build:css && pnpm build:api",
+    "prepare:py-static": "pnpm prepare:static && pnpm build:static && pnpm copy:static",
+    "copy:next-src": "cp -r ./next ./server",
+    "build:css": "postcss server/app/globals.css -o server/app/globals.css && rm -rf ./server/postcss.config.js",
+    "build:api": "rm -rf ./server/app/api && tsc --skipLibCheck --project tsconfig.api.json",
+    "prepare:static": "cp -r ./next ./temp && rm -rf ./temp/app/api && mv ./temp/next-build.config.ts ./temp/next.config.ts",
+    "build:static": "cd ./temp && next build",
+    "copy:static": "cp -r ./temp/out ./dist/static && rm -rf ./temp"
   },
   "devDependencies": {
     "@tailwindcss/postcss": "^4",
     "@types/babel__standalone": "^7.1.9",
     "@types/babel__traverse": "^7.20.7",
-    "llamaindex": "0.10.2",
+    "llamaindex": "~0.11.0",
     "postcss": "^8.5.3",
     "postcss-cli": "^11.0.1",
     "tailwindcss": "^4",
@@ -103,9 +107,9 @@
     "vaul": "^1.1.2"
   },
   "peerDependencies": {
-    "@llamaindex/env": "^0.1.29",
-    "@llamaindex/workflow": "^1.1.0",
-    "llamaindex": "^0.10.2",
+    "@llamaindex/env": "~0.1.30",
+    "@llamaindex/workflow": "~1.1.3",
+    "llamaindex": "~0.11.0",
     "zod": "^3.24.2",
     "zod-to-json-schema": "^3.23.3"
   },

packages/server/src/events.ts

@@ -155,10 +155,12 @@ export function extractAllArtifacts(messages: Message[]): Artifact[] {
     const artifacts =
       message.annotations
         ?.filter(
-          (annotation) =>
+          (
+            annotation,
+          ): annotation is z.infer<typeof artifactAnnotationSchema> =>
             artifactAnnotationSchema.safeParse(annotation).success,
         )
-        .map((artifact) => artifact as Artifact) ?? [];
+        .map((annotation) => annotation.data as Artifact) ?? [];
 
     allArtifacts.push(...artifacts);
   }

packages/server/src/handlers/cloud.ts

@@ -1,30 +0,0 @@
-import { getEnv } from "@llamaindex/env";
-import type { IncomingMessage, ServerResponse } from "http";
-import { LLamaCloudFileService } from "llamaindex";
-import { sendJSONResponse } from "../utils/request";
-
-export const getLlamaCloudConfig = async (
-  req: IncomingMessage,
-  res: ServerResponse,
-) => {
-  if (!getEnv("LLAMA_CLOUD_API_KEY")) {
-    return sendJSONResponse(res, 500, {
-      error: "env variable LLAMA_CLOUD_API_KEY is required to use LlamaCloud",
-    });
-  }
-
-  try {
-    const config = {
-      projects: await LLamaCloudFileService.getAllProjectsWithPipelines(),
-      pipeline: {
-        pipeline: getEnv("LLAMA_CLOUD_INDEX_NAME"),
-        project: getEnv("LLAMA_CLOUD_PROJECT_NAME"),
-      },
-    };
-    return sendJSONResponse(res, 200, config);
-  } catch (error) {
-    return sendJSONResponse(res, 500, {
-      error: "Failed to fetch LlamaCloud configuration",
-    });
-  }
-};

packages/server/src/handlers/files.ts

@@ -1,106 +0,0 @@
-import { exec } from "child_process";
-import fs from "fs";
-import type { IncomingMessage, ServerResponse } from "http";
-import path from "path";
-import { promisify } from "util";
-import { parseRequestBody, sendJSONResponse } from "../utils/request";
-
-export const handleServeFiles = async (
-  req: IncomingMessage,
-  res: ServerResponse,
-  pathname: string,
-) => {
-  const filePath = pathname.substring("/api/files/".length);
-  if (!filePath.startsWith("output") && !filePath.startsWith("data")) {
-    return sendJSONResponse(res, 400, { error: "No permission" });
-  }
-  const decodedFilePath = decodeURIComponent(filePath);
-  const fileExists = await promisify(fs.exists)(decodedFilePath);
-  if (fileExists) {
-    const fileStream = fs.createReadStream(decodedFilePath);
-    fileStream.pipe(res);
-  } else {
-    return sendJSONResponse(res, 404, { error: "File not found" });
-  }
-};
-
-const DEFAULT_WORKFLOW_FILE_PATH = "src/app/workflow.ts"; // TODO: we can make it as a parameter in server later
-
-export const getWorkflowFile = async (
-  req: IncomingMessage,
-  res: ServerResponse,
-  filePath: string = DEFAULT_WORKFLOW_FILE_PATH,
-) => {
-  const fileExists = await promisify(fs.exists)(filePath);
-  if (!fileExists) {
-    return sendJSONResponse(res, 404, {
-      detail: `Dev mode is currently in beta. It only supports updating workflow file at ${DEFAULT_WORKFLOW_FILE_PATH}`,
-    });
-  }
-
-  const content = await promisify(fs.readFile)(filePath, "utf-8");
-  const last_modified = fs.statSync(filePath).mtime.getTime();
-  sendJSONResponse(res, 200, { content, file_path: filePath, last_modified });
-};
-
-export const updateWorkflowFile = async (
-  req: IncomingMessage,
-  res: ServerResponse,
-  filePath: string = DEFAULT_WORKFLOW_FILE_PATH,
-) => {
-  const body = await parseRequestBody(req);
-  const { content } = body as { content: string };
-
-  const fileExists = await promisify(fs.exists)(filePath);
-  if (!fileExists) {
-    return sendJSONResponse(res, 404, {
-      detail: `Dev mode is currently in beta. It only supports updating workflow file at ${DEFAULT_WORKFLOW_FILE_PATH}`,
-    });
-  }
-
-  try {
-    const resolvedFilePath = path.resolve(DEFAULT_WORKFLOW_FILE_PATH);
-    const result = await validateTypeScriptFile(resolvedFilePath, content);
-
-    if (!result.isValid) {
-      return sendJSONResponse(res, 400, {
-        detail: result.errors.join("\n"),
-      });
-    }
-
-    await promisify(fs.writeFile)(filePath, content);
-    sendJSONResponse(res, 200, { content });
-  } catch (error) {
-    console.error("Error updating workflow file:", error);
-    sendJSONResponse(res, 500, { error: "Failed to update workflow file" });
-  }
-};
-
-// use typescript package to validate the file syntax and imports
-async function validateTypeScriptFile(filePath: string, content: string) {
-  // Update workflow file directly will cause the server restart immediately.
-  // So we create a temporary file with the same content in the same directory as the workflow file
-  // This file will be used to validate the file syntax and imports. It will be deleted after validation.
-  const tempFilePath = path.join(
-    path.dirname(filePath),
-    `workflow_${Date.now()}.ts`,
-  );
-  fs.writeFileSync(tempFilePath, content);
-
-  const errors = [];
-  try {
-    const tscCommand = `npx tsc ${tempFilePath} --noEmit --skipLibCheck true`;
-    await promisify(exec)(tscCommand);
-  } catch (error) {
-    const errorMessage = (error as { stdout: string })?.stdout;
-    errors.push(errorMessage);
-  } finally {
-    // Clean up temporary file
-    if (fs.existsSync(tempFilePath)) fs.unlinkSync(tempFilePath);
-  }
-
-  return {
-    isValid: errors.length === 0,
-    errors: errors,
-  };
-}

packages/server/src/server.ts

@@ -7,13 +7,6 @@ import path from "path";
 import { parse } from "url";
 import { promisify } from "util";
 import { handleChat } from "./handlers/chat";
-import { getLlamaCloudConfig } from "./handlers/cloud";
-import { getComponents } from "./handlers/components";
-import {
-  getWorkflowFile,
-  handleServeFiles,
-  updateWorkflowFile,
-} from "./handlers/files";
 import type { LlamaIndexServerOptions } from "./types";
 
 const nextDir = path.join(__dirname, "..", "server");
@@ -78,41 +71,25 @@ export class LlamaIndexServer {
     const server = createServer((req, res) => {
       const parsedUrl = parse(req.url!, true);
       const pathname = parsedUrl.pathname;
+      const query = parsedUrl.query;
 
       if (pathname === "/api/chat" && req.method === "POST") {
+        // because of https://github.com/vercel/next.js/discussions/79402 we can't use route.ts here, so we need to call this custom route
+        // when calling `pnpm eject`, the user will get an equivalent route at [path to chat route.ts]
+        // make sure to keep its semantic in sync with handleChat
         return handleChat(req, res, this.workflowFactory);
       }
 
-      if (pathname?.startsWith("/api/files") && req.method === "GET") {
-        return handleServeFiles(req, res, pathname);
-      }
-
       if (
         this.componentsDir &&
         pathname === "/api/components" &&
         req.method === "GET"
       ) {
-        return getComponents(req, res, this.componentsDir);
-      }
-
-      if (
-        getEnv("LLAMA_CLOUD_API_KEY") &&
-        pathname === "/api/chat/config/llamacloud" &&
-        req.method === "GET"
-      ) {
-        return getLlamaCloudConfig(req, res);
-      }
-
-      if (pathname === "/api/dev/files/workflow" && req.method === "GET") {
-        return getWorkflowFile(req, res);
-      }
-
-      if (pathname === "/api/dev/files/workflow" && req.method === "PUT") {
-        return updateWorkflowFile(req, res);
+        query.componentsDir = this.componentsDir;
       }
 
       const handle = this.app.getRequestHandler();
-      handle(req, res, parsedUrl);
+      handle(req, res, { ...parsedUrl, query });
     });
 
     server.listen(this.port, () => {

packages/server/tsconfig.api.json

@@ -0,0 +1,10 @@
+{
+  "extends": "./tsconfig.json",
+  "compilerOptions": {
+    "rootDir": "./next/app/api",
+    "outDir": "./server/app/api",
+    "emitDeclarationOnly": false
+  },
+  "include": ["./next/app/api"],
+  "exclude": ["./next/app/api/chat/route.ts"]
+}

pnpm-lock.yaml

@@ -184,11 +184,11 @@ importers:
         specifier: 0.4.4
         version: 0.4.4(@babel/runtime@7.27.0)(@codemirror/autocomplete@6.18.6)(@codemirror/language@6.11.0)(@codemirror/lint@6.8.5)(@codemirror/search@6.5.10)(@codemirror/state@6.5.2)(@codemirror/theme-one-dark@6.1.2)(@codemirror/view@6.36.7)(@types/react-dom@19.1.2(@types/react@19.1.2))(@types/react@19.1.2)(codemirror@6.0.1)(react-dom@19.1.0(react@19.1.0))(react@19.1.0)
       '@llamaindex/env':
-        specifier: ^0.1.29
-        version: 0.1.29
+        specifier: ~0.1.30
+        version: 0.1.30
       '@llamaindex/workflow':
-        specifier: ^1.1.0
-        version: 1.1.0(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(zod@3.24.3)
+        specifier: ~1.1.3
+        version: 1.1.3(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)
       '@radix-ui/react-accordion':
         specifier: ^1.2.3
         version: 1.2.8(@types/react-dom@19.1.2(@types/react@19.1.2))(@types/react@19.1.2)(react-dom@19.1.0(react@19.1.0))(react@19.1.0)
@@ -341,8 +341,8 @@ importers:
         specifier: ^7.20.7
         version: 7.20.7
       llamaindex:
-        specifier: 0.10.2
-        version: 0.10.2(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)(zod@3.24.3)
+        specifier: ~0.11.0
+        version: 0.11.1(@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3))(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)(zod@3.24.3)
       postcss:
         specifier: ^8.5.3
         version: 8.5.3
@@ -378,7 +378,7 @@ importers:
         version: 0.0.4(openapi-types@12.1.3)
       '@llamaindex/workflow':
         specifier: 1.1.0
-        version: 1.1.0(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(zod@3.24.3)
+        version: 1.1.0(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)
       dotenv:
         specifier: ^16.4.7
         version: 16.5.0
@@ -1191,12 +1191,22 @@ packages:
       '@llamaindex/core': 0.6.2
       '@llamaindex/env': 0.1.29
 
+  '@llamaindex/cloud@4.0.9':
+    resolution: {integrity: sha512-+tIPghKAzXFvKEPmbJZpcFJgzmxLq/HfSKlJAIinRdrSDBC489oAJsqkt6StxmlFUhWS3ATOtTWpqWZIQ9HtzA==}
+    peerDependencies:
+      '@llama-flow/core': ^0.4.1
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+
   '@llamaindex/core@0.6.1':
     resolution: {integrity: sha512-OsY8JV2nbfWYjp24HEKDbtM6vOa0TNUreMjDMxUkw5Wet1TxKxjVq0YHY01RoqMbbTJ+Jsxa42FhzTaDV2gMkg==}
 
   '@llamaindex/core@0.6.2':
     resolution: {integrity: sha512-e0NH7X1C8yq9sCqj1Ys9jL9rITCXBKfZwmKnTHYxpXorPeuNmlFmuoogahJTwLNsILBDTOVAUexO88SqmAFUrA==}
 
+  '@llamaindex/core@0.6.6':
+    resolution: {integrity: sha512-UplAAxsv3LnU1EP2kPuyqDtRQmQqYtsI00rVDeexOwgqwNTfGddI0yPRrFTBSNe0q2ttBuPsB7D3M7j4UY3yXQ==}
+
   '@llamaindex/env@0.1.29':
     resolution: {integrity: sha512-GoITt+QLDNIhu2i1sGsPH8tHH13Gxp4i4ofMFlefrHagytvF761MG7nvTAQVIqP9kxBYk4dnSQ3lQnVPFAfMZQ==}
     peerDependencies:
@@ -1208,6 +1218,17 @@ packages:
       gpt-tokenizer:
         optional: true
 
+  '@llamaindex/env@0.1.30':
+    resolution: {integrity: sha512-y6kutMcCevzbmexUgz+HXf7KiZemzAoFEYSjAILfR+cG6FmYSF8XvLbGOB34Kx8mlRi7EI8rZXpezJ5qCqOyZg==}
+    peerDependencies:
+      '@huggingface/transformers': ^3.5.0
+      gpt-tokenizer: ^2.5.0
+    peerDependenciesMeta:
+      '@huggingface/transformers':
+        optional: true
+      gpt-tokenizer:
+        optional: true
+
   '@llamaindex/node-parser@2.0.2':
     resolution: {integrity: sha512-pWqEfFRMW5TFzbDyEDWqhipR9P/HBF2C+pEeTroyRACZtKJLbeg3R3e2g4gmfnPxmxQIXkGn2UFK8Io4nwVNOg==}
     peerDependencies:
@@ -1216,6 +1237,14 @@ packages:
       tree-sitter: ^0.22.0
       web-tree-sitter: ^0.24.3
 
+  '@llamaindex/node-parser@2.0.6':
+    resolution: {integrity: sha512-8vJ7hsh1FKfAfO90nNe/G+/E2hM31dT/+1+0cPpZp02Ez98Jhwj+vTcCRlvyZznCn6DzHNBuw+uP1wKCqylSnQ==}
+    peerDependencies:
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      tree-sitter: ^0.22.0
+      web-tree-sitter: ^0.24.3
+
   '@llamaindex/openai@0.2.1':
     resolution: {integrity: sha512-gSBYL1vMaEihXm5K5gIHpUmvDS78ACPgQPlwf/ZnE82UD/vFNOWOqQbwESJKcygC0lHrwhrKJAuBAzXSNcpznA==}
 
@@ -1255,6 +1284,13 @@ packages:
       '@llamaindex/env': 0.1.29
       zod: ^3.23.8
 
+  '@llamaindex/workflow@1.1.3':
+    resolution: {integrity: sha512-B9wlN4xMtfbnIQW84wTC0GdyIaLGXYnZ4I0LENV2UM/0wqlXKI7YoBTO3cJvi6pO1sPA+K29Tqw822VsGF+GkA==}
+    peerDependencies:
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      zod: ^3.23.8
+
   '@manypkg/find-root@1.1.0':
     resolution: {integrity: sha512-mki5uBvhHzO8kYYix/WRy2WX8S3B5wdVSc9D6KcU5lQNglP2yt58/VfLuAK49glRXChosY8ap2oJ1qgma3GUVA==}
 
@@ -2484,6 +2520,9 @@ packages:
   '@types/responselike@1.0.3':
     resolution: {integrity: sha512-H/+L+UkTV33uf49PH5pCAUBVPNj2nDBXTN+qS1dOwyyg24l3CcicicCA7ca+HMvJBZcFgl5r8e+RR6elsb4Lyw==}
 
+  '@types/retry@0.12.2':
+    resolution: {integrity: sha512-XISRgDJ2Tc5q4TRqvgJtzsRkFYNJzZrhTdtMoGVBttwzzQJkPnS3WWTFc7kuDRoPtPakl+T+OfdEUjYJj7Jbow==}
+
   '@types/retry@0.12.5':
     resolution: {integrity: sha512-3xSjTp3v03X/lSQLkczaN9UIEwJMoMCA1+Nb5HfbJEQWogdeQIyVtTvxPXDQjZ5zws8rFQfVfRdz03ARihPJgw==}
 
@@ -4127,6 +4166,10 @@ packages:
   is-module@1.0.0:
     resolution: {integrity: sha512-51ypPSPCoTEIN9dy5Oy+h4pShgJmPCygKfyRCISBI+JoWT/2oJvK8QPxmwv7b/p239jXrm9M1mlQbyKJ5A152g==}
 
+  is-network-error@1.1.0:
+    resolution: {integrity: sha512-tUdRRAnhT+OtCZR/LxZelH/C7QtjtFrTu5tXCA8pl55eTUElUHT+GPYV8MBMBvea/j+NxQqVt3LbWMRir7Gx9g==}
+    engines: {node: '>=16'}
+
   is-number-object@1.1.1:
     resolution: {integrity: sha512-lZhclumE1G6VYD8VHe35wFaIif+CTy5SJIi5+3y4psDgWu4wPDoBhF8NxUOinEc7pHgiTsT6MaBb92rKhhD+Xw==}
     engines: {node: '>= 0.4'}
@@ -4397,6 +4440,10 @@ packages:
     resolution: {integrity: sha512-tUbb+HYaQ+WP1f4NrYnEOIgPPRUePh3sPnOf/wbxTYAyZTnREpSiwCfLUGgYOiIYKK5PS7Oo/eTZCP7jpi7gSA==}
     engines: {node: '>=18.0.0'}
 
+  llamaindex@0.11.1:
+    resolution: {integrity: sha512-d+sMHc6b/Lw8S1HhTVtR5hgEkg4uCpADpgZDcpHa/QlSQXvCmSkEdI9DOhoTDMhNWWatKnQZADPDuVzkyysQ3w==}
+    engines: {node: '>=18.0.0'}
+
   locate-path@5.0.0:
     resolution: {integrity: sha512-t7hw9pI+WvuwNJXwk5zVHpyhIqzg2qTlklJOf0mVxGSbe3Fp2VieZcduNYjaLDoy6p9uGpQEGWG87WpMKlNq8g==}
     engines: {node: '>=8'}
@@ -4977,6 +5024,10 @@ packages:
     resolution: {integrity: sha512-y3b8Kpd8OAN444hxfBbFfj1FY/RjtTd8tzYwhUqNYXx0fXx2iX4maP4Qr6qhIKbQXI02wTLAda4fYUbDagTUFw==}
     engines: {node: '>=6'}
 
+  p-retry@6.2.1:
+    resolution: {integrity: sha512-hEt02O4hUct5wtwg4H4KcWgDdm+l1bOaEy/hWzd8xtXB9BqxTWBBhb+2ImAtH4Cv4rPjV76xN3Zumqk3k3AhhQ==}
+    engines: {node: '>=16.17'}
+
   p-timeout@3.2.0:
     resolution: {integrity: sha512-rhIwUycgwwKcP9yTOOFK/AKsAopjjCakVqLHePO3CC6Mir1Z99xT+R63jZxAT5lFZLa2inS5h+ZS2GvR99/FBg==}
     engines: {node: '>=8'}
@@ -5521,6 +5572,10 @@ packages:
     resolution: {integrity: sha512-9LkiTwjUh6rT555DtE9rTX+BKByPfrMzEAtnlEtdEwr3Nkffwiihqe2bWADg+OQRjt9gl6ICdmB/ZFDCGAtSow==}
     engines: {node: '>= 4'}
 
+  retry@0.13.1:
+    resolution: {integrity: sha512-XQBQ3I8W1Cge0Seh+6gjj03LbmRFWuoszgK9ooCpwYIrhhoO80pfq4cUkU5DkknwfOfFteRwlZ56PYOGYyFWdg==}
+    engines: {node: '>= 4'}
+
   reusify@1.1.0:
     resolution: {integrity: sha512-g6QUff04oZpHs0eG5p83rFLhHeV00ug/Yf9nZM6fLeUrPguBTkTQOdpAWWspMh55TZfVQDPaN3NQJfbVRAxdIw==}
     engines: {iojs: '>=1.0.0', node: '>=0.10.0'}
@@ -6381,6 +6436,9 @@ packages:
   zod@3.24.3:
     resolution: {integrity: sha512-HhY1oqzWCQWuUqvBFnsyrtZRhyPeR7SUGv+C4+MsisMuVfSPx8HpwWqH8tRahSlt6M3PiFAcoeFhZAqIXTxoSg==}
 
+  zod@3.25.13:
+    resolution: {integrity: sha512-Q8mvk2iWi7rTDfpQBsu4ziE7A6AxgzJ5hzRyRYQkoV3A3niYsXVwDaP1Kbz3nWav6S+VZ6k2OznFn8ZyDHvIrg==}
+
   zwitch@2.0.4:
     resolution: {integrity: sha512-bXE4cR/kVZhKZX/RjPEflHaKVhUVl85noU3v6b8apfQEc1x4A+zBxjZ4lN8LqGd6WZ3dl98pY4o717VFmoPp+A==}
 
@@ -7141,9 +7199,10 @@ snapshots:
       '@lezer/highlight': 1.2.1
       '@lezer/lr': 1.4.2
 
-  '@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(zod@3.24.3)':
+  '@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)':
     optionalDependencies:
       next: 15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0)
+      p-retry: 6.2.1
       zod: 3.24.3
 
   '@llamaindex/chat-ui@0.4.4(@babel/runtime@7.27.0)(@codemirror/autocomplete@6.18.6)(@codemirror/language@6.11.0)(@codemirror/lint@6.8.5)(@codemirror/search@6.5.10)(@codemirror/state@6.5.2)(@codemirror/theme-one-dark@6.1.2)(@codemirror/view@6.36.7)(@types/react-dom@19.1.2(@types/react@19.1.2))(@types/react@19.1.2)(codemirror@6.0.1)(react-dom@19.1.0(react@19.1.0))(react@19.1.0)':
@@ -7206,6 +7265,14 @@ snapshots:
       '@llamaindex/core': 0.6.2
       '@llamaindex/env': 0.1.29
 
+  '@llamaindex/cloud@4.0.9(@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3))(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)':
+    dependencies:
+      '@llama-flow/core': 0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      p-retry: 6.2.1
+      zod: 3.25.13
+
   '@llamaindex/core@0.6.1':
     dependencies:
       '@llamaindex/env': 0.1.29
@@ -7228,12 +7295,29 @@ snapshots:
       - '@huggingface/transformers'
       - gpt-tokenizer
 
+  '@llamaindex/core@0.6.6':
+    dependencies:
+      '@llamaindex/env': 0.1.30
+      '@types/node': 22.15.2
+      magic-bytes.js: 1.12.1
+      zod: 3.24.3
+      zod-to-json-schema: 3.24.5(zod@3.24.3)
+    transitivePeerDependencies:
+      - '@huggingface/transformers'
+      - gpt-tokenizer
+
   '@llamaindex/env@0.1.29':
     dependencies:
       '@aws-crypto/sha256-js': 5.2.0
       js-tiktoken: 1.0.20
       pathe: 1.1.2
 
+  '@llamaindex/env@0.1.30':
+    dependencies:
+      '@aws-crypto/sha256-js': 5.2.0
+      js-tiktoken: 1.0.20
+      pathe: 1.1.2
+
   '@llamaindex/node-parser@2.0.2(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)':
     dependencies:
       '@llamaindex/core': 0.6.2
@@ -7242,6 +7326,14 @@ snapshots:
       tree-sitter: 0.22.4
       web-tree-sitter: 0.24.7
 
+  '@llamaindex/node-parser@2.0.6(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)':
+    dependencies:
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      html-to-text: 9.0.5
+      tree-sitter: 0.22.4
+      web-tree-sitter: 0.24.7
+
   '@llamaindex/openai@0.2.1':
     dependencies:
       '@llamaindex/core': 0.6.1
@@ -7319,9 +7411,15 @@ snapshots:
       '@llamaindex/env': 0.1.29
       zod: 3.24.3
 
-  '@llamaindex/workflow@1.1.0(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(zod@3.24.3)':
+  '@llamaindex/workflow@1.0.3(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(zod@3.24.3)':
     dependencies:
-      '@llama-flow/core': 0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(zod@3.24.3)
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      zod: 3.24.3
+
+  '@llamaindex/workflow@1.1.0(@llamaindex/core@0.6.2)(@llamaindex/env@0.1.29)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)':
+    dependencies:
+      '@llama-flow/core': 0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)
       '@llamaindex/core': 0.6.2
       '@llamaindex/env': 0.1.29
       zod: 3.24.3
@@ -7332,6 +7430,19 @@ snapshots:
       - p-retry
       - rxjs
 
+  '@llamaindex/workflow@1.1.3(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)':
+    dependencies:
+      '@llama-flow/core': 0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3)
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      zod: 3.24.3
+    transitivePeerDependencies:
+      - '@modelcontextprotocol/sdk'
+      - hono
+      - next
+      - p-retry
+      - rxjs
+
   '@manypkg/find-root@1.1.0':
     dependencies:
       '@babel/runtime': 7.27.0
@@ -8528,6 +8639,8 @@ snapshots:
     dependencies:
       '@types/node': 20.17.30
 
+  '@types/retry@0.12.2': {}
+
   '@types/retry@0.12.5': {}
 
   '@types/tar@6.1.5':
@@ -10467,6 +10580,8 @@ snapshots:
 
   is-module@1.0.0: {}
 
+  is-network-error@1.1.0: {}
+
   is-number-object@1.1.1:
     dependencies:
       call-bound: 1.0.4
@@ -10745,6 +10860,26 @@ snapshots:
       - ws
       - zod
 
+  llamaindex@0.11.1(@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3))(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)(zod@3.24.3):
+    dependencies:
+      '@llamaindex/cloud': 4.0.9(@llama-flow/core@0.4.1(next@15.3.1(@opentelemetry/api@1.9.0)(@playwright/test@1.52.0)(react-dom@19.1.0(react@19.1.0))(react@19.1.0))(p-retry@6.2.1)(zod@3.24.3))(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)
+      '@llamaindex/core': 0.6.6
+      '@llamaindex/env': 0.1.30
+      '@llamaindex/node-parser': 2.0.6(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(tree-sitter@0.22.4)(web-tree-sitter@0.24.7)
+      '@llamaindex/workflow': 1.0.3(@llamaindex/core@0.6.6)(@llamaindex/env@0.1.30)(zod@3.24.3)
+      '@types/lodash': 4.17.16
+      '@types/node': 22.15.2
+      ajv: 8.17.1
+      lodash: 4.17.21
+      magic-bytes.js: 1.12.1
+    transitivePeerDependencies:
+      - '@huggingface/transformers'
+      - '@llama-flow/core'
+      - gpt-tokenizer
+      - tree-sitter
+      - web-tree-sitter
+      - zod
+
   locate-path@5.0.0:
     dependencies:
       p-locate: 4.1.0
@@ -11482,6 +11617,12 @@ snapshots:
 
   p-map@2.1.0: {}
 
+  p-retry@6.2.1:
+    dependencies:
+      '@types/retry': 0.12.2
+      is-network-error: 1.1.0
+      retry: 0.13.1
+
   p-timeout@3.2.0:
     dependencies:
       p-finally: 1.0.0
@@ -12046,6 +12187,8 @@ snapshots:
 
   retry@0.12.0: {}
 
+  retry@0.13.1: {}
+
   reusify@1.1.0: {}
 
   rfdc@1.4.1: {}
@@ -13089,4 +13232,6 @@ snapshots:
 
   zod@3.24.3: {}
 
+  zod@3.25.13: {}
+
   zwitch@2.0.4: {}

python/llama-index-server/README.md

@@ -72,7 +72,7 @@ app = LlamaIndexServer(
 
 The LlamaIndexServer accepts the following configuration parameters:
 
-- `workflow_factory`: A callable that creates a workflow instance for each request
+- `workflow_factory`: A callable that creates a workflow instance for each request. See [Workflow factory contract](#workflow-factory-contract) for more details.
 - `logger`: Optional logger instance (defaults to uvicorn logger)
 - `use_default_routers`: Whether to include default routers (chat, static file serving)
 - `env`: Environment setting ('dev' enables CORS and UI by default)
@@ -88,6 +88,31 @@ The LlamaIndexServer accepts the following configuration parameters:
 - `api_prefix`: API route prefix (default: "/api")
 - `server_url`: The deployment URL of the server (default is None)
 
+## Workflow factory contract
+
+The `workflow_factory` provided will be called for each chat request to initialize a new workflow instance. Additionally, we provide the [ChatRequest](https://github.com/run-llama/create-llama/blob/afe9e9fc16427d20e1dfb635a45e7ed4b46285cb/python/llama-index-server/llama_index/server/api/models.py#L32) object, which includes the request information that is helpful for initializing the workflow. For example:
+```python
+def create_workflow(chat_request: ChatRequest) -> Workflow:
+    # using messages from the chat request to initialize the workflow
+    return MyCustomWorkflow(chat_request.messages)
+```
+
+Your workflow will be executed once for each chat request with the following input parameters are included in workflow's `StartEvent`:
+- `user_msg` [str]: The current user message
+- `chat_history` [list[[ChatMessage](https://docs.llamaindex.ai/en/stable/api_reference/prompts/#llama_index.core.prompts.ChatMessage)]]: All the previous messages of the conversation
+
+Example:
+```python
+@step
+def handle_start_event(ev: StartEvent) -> MyNextEvent:
+    user_msg = ev.user_msg
+    chat_history = ev.chat_history
+    ...
+```
+
+Your workflows can emit `UIEvent` events to render [Custom UI Components](https://github.com/run-llama/create-llama/blob/main/python/llama-index-server/docs/custom_ui_component.md) in the chat UI to improve the user experience.
+Furthermore, you can send `ArtifactEvent` events to render code or document [Artifacts](https://github.com/run-llama/create-llama/blob/main/python/llama-index-server/docs/custom_artifact_event.md) in a dedicated Canvas panel in the chat UI.
+
 ## Default Routers and Features
 
 ### Chat Router
@@ -108,11 +133,6 @@ When enabled, the server provides a chat interface at the root path (`/`) with:
 - Real-time chat interface
 - API endpoint integration
 
-### Custom UI Components
-
-You can add custom UI components for your workflow by providing `component_dir` config and adding custom .jsx or .tsx files to the directory.
-See [Custom UI Components](https://github.com/run-llama/create-llama/blob/main/llama-index-server/docs/custom_ui_component.md) for more details.
-
 ## Development Mode
 
 In development mode (`env="dev"`), the server:
@@ -135,7 +155,6 @@ app = LlamaIndexServer(
 
 **Note**: The workflow editor is currently in beta and only supports updating LlamaIndexServer projects created with [create-llama](https://github.com/run-llama/create-llama/). You also need to start the server via `fastapi dev` so that the server can hot reload the workflow code.
 
-
 ## API Endpoints
 
 The server provides the following default endpoints:
@@ -146,11 +165,10 @@ The server provides the following default endpoints:
 
 ## Best Practices
 
-1. Always provide a workflow factory that creates fresh workflow instances
-2. Use environment variables for sensitive configuration
-3. Enable verbose logging during development
-4. Configure CORS appropriately for your deployment environment
-5. Use starter questions to guide users in the chat UI
+1. Use environment variables for sensitive configuration
+2. Enable verbose logging during development
+3. Configure CORS appropriately for your deployment environment
+4. Use starter questions to guide users in the chat UI
 
 ## Getting Started with a New Project
 

python/llama-index-server/docs/custom_artifact_event.md

@@ -0,0 +1,59 @@
+# Sending Artifacts to the UI
+
+In addition to UI events for custom components, LlamaIndex Server supports a special `ArtifactEvent` to send structured data like generated documents or code snippets to the UI. These artifacts are displayed in a dedicated "Canvas" panel in the chat interface.
+
+## Artifact Event Structure
+
+To send an artifact, your workflow needs to emit an event with `type: "artifact"`. The `data` payload of this event should include:
+
+- `type`: An `ArtifactType` enum indicating the type of artifact (e.g., `ArtifactType.DOCUMENT`, `ArtifactType.CODE`).
+- `created_at`: A timestamp (e.g., `int(time.time())`) indicating when the artifact was created.
+- `data`: An object containing the specific details of the artifact. The structure of this object depends on the artifact `type`. For example, `DocumentArtifactData` or `CodeArtifactData`.
+
+## Defining and Sending an ArtifactEvent
+
+First, import the necessary classes:
+
+```python
+import time
+from llama_index.server.api.models import (
+    Artifact,
+    ArtifactEvent,
+    ArtifactType,
+    DocumentArtifactData,
+    # CodeArtifactData, # Import if sending code artifacts
+)
+```
+
+Then, within your workflow logic, use `ctx.write_event_to_stream` to emit the event. Here's an example of sending a document artifact, taken from [document_workflow.py](/python/llama-index-server/examples/artifact/document_workflow.py):
+
+```python
+# Assuming 'ctx' is the workflow Context and 'content' is a markdown string
+
+ctx.write_event_to_stream(
+    ArtifactEvent(
+        data=Artifact(
+            type=ArtifactType.DOCUMENT,
+            created_at=int(time.time()),
+            data=DocumentArtifactData(
+                title="My cooking recipes",
+                content=content,
+                type="markdown",
+            ),
+        ),
+    )
+)
+```
+
+This will send the artifact to the LlamaIndex Server UI, where it will be rendered in the Canvas panel by a renderer depending on the artifact type. For `ArtifactType.DOCUMENT`, this uses a `DocumentArtifactViewer`.
+
+## Available Artifact Types
+
+LlamaIndex Server currently supports the following artifact types:
+
+- `ArtifactType.DOCUMENT`: For text-based documents like Markdown or HTML.
+  - `data` should be an instance of `DocumentArtifactData` which includes `title: str`, `content: str`, and `type: Literal["markdown", "html"]`.
+- `ArtifactType.CODE`: For code snippets.
+  - `data` should be an instance of `CodeArtifactData` which includes `title: str`, `code: str`, and `language: str`.
+
+Ensure you provide the correct data model corresponding to the `ArtifactType` you are sending. You can find these data models in `llama_index.server.api.models`. 

python/llama-index-server/llama_index/server/chat_ui.py

@@ -5,7 +5,7 @@ from typing import Optional
 
 import requests
 
-CHAT_UI_VERSION = "0.1.6"
+CHAT_UI_VERSION = "0.2.1"
 
 
 def download_chat_ui(

python/llama-index-server/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "llama-index-server"
-version = "0.1.15"
+version = "0.1.16"
 description = "llama-index fastapi server"
 readme = "README.md"
 license = "MIT"