add tool call events for agent_tool utils

.changeset/mean-fireants-visit.md

@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+fix: add trycatch for generating error

.changeset/rare-eyes-protect.md

@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+bump: chat-ui and tailwind v4

.eslintrc

@@ -0,0 +1,12 @@
+{
+  "extends": [
+    "prettier"
+  ],
+  "rules": {
+    "max-params": [
+      "error",
+      4
+    ],
+    "prefer-const": "error",
+  },
+}

.github/workflows/e2e.yml

@@ -1,15 +1,12 @@
-name: E2E Tests for create-llama package
+name: E2E Tests
 on:
   push:
     branches: [main]
-    paths-ignore:
-      - "python/llama-index-server/**"
-      - ".github/workflows/*llama_index_server.yml"
   pull_request:
     branches: [main]
-    paths-ignore:
-      - "python/llama-index-server/**"
-      - ".github/workflows/*llama_index_server.yml"
+
+env:
+  POETRY_VERSION: "1.6.1"
 
 jobs:
   e2e-python:
@@ -35,10 +32,10 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
 
-      - name: Install uv
-        run: curl -LsSf https://astral.sh/uv/install.sh | sh
-      - name: Add uv to PATH # Ensure uv is available in subsequent steps
-        run: echo "$HOME/.cargo/bin" >> $GITHUB_PATH
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          version: ${{ env.POETRY_VERSION }}
 
       - uses: pnpm/action-setup@v3
 
@@ -53,15 +50,15 @@ jobs:
 
       - name: Install Playwright Browsers
         run: pnpm exec playwright install --with-deps
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Build create-llama
         run: pnpm run build
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Install
         run: pnpm run pack-install
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Run Playwright tests for Python
         run: pnpm run e2e:python
@@ -70,15 +67,13 @@ jobs:
           LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
           FRAMEWORK: ${{ matrix.frameworks }}
           DATASOURCE: ${{ matrix.datasources }}
-          PYTHONIOENCODING: utf-8
-          PYTHONLEGACYWINDOWSSTDIO: utf-8
-        working-directory: packages/create-llama
+        working-directory: .
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
           name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}
-          path: packages/create-llama/playwright-report/
+          path: ./playwright-report/
           overwrite: true
           retention-days: 30
 
@@ -88,10 +83,10 @@ jobs:
     strategy:
       fail-fast: true
       matrix:
-        node-version: [20, 22]
+        node-version: [18, 20]
         python-version: ["3.11"]
         os: [macos-latest, windows-latest, ubuntu-22.04]
-        frameworks: ["nextjs"]
+        frameworks: ["nextjs", "express"]
         datasources: ["--no-files", "--example-file", "--llamacloud"]
     defaults:
       run:
@@ -105,10 +100,10 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
 
-      - name: Install uv
-        run: curl -LsSf https://astral.sh/uv/install.sh | sh
-      - name: Add uv to PATH # Ensure uv is available in subsequent steps
-        run: echo "$HOME/.cargo/bin" >> $GITHUB_PATH
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          version: ${{ env.POETRY_VERSION }}
 
       - uses: pnpm/action-setup@v3
 
@@ -123,15 +118,15 @@ jobs:
 
       - name: Install Playwright Browsers
         run: pnpm exec playwright install --with-deps
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Build create-llama
         run: pnpm run build
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Install
         run: pnpm run pack-install
-        working-directory: packages/create-llama
+        working-directory: .
 
       - name: Run Playwright tests for TypeScript
         run: pnpm run e2e:typescript
@@ -140,12 +135,12 @@ jobs:
           LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
           FRAMEWORK: ${{ matrix.frameworks }}
           DATASOURCE: ${{ matrix.datasources }}
-        working-directory: packages/create-llama
+        working-directory: .
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
           name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}
-          path: packages/create-llama/playwright-report/
+          path: ./playwright-report/
           overwrite: true
           retention-days: 30

.github/workflows/lint_on_push_or_pull.yml

@@ -35,10 +35,8 @@ jobs:
         uses: chartboost/ruff-action@v1
         with:
           args: "format --check"
-          src: "python/llama-index-server"
 
       - name: Run Python lint
         uses: chartboost/ruff-action@v1
         with:
           args: "check"
-          src: "python/llama-index-server"

.github/workflows/release_llama_index_server.yml

@@ -5,7 +5,7 @@ on:
     branches:
       - main
     paths:
-      - "python/llama-index-server/**"
+      - "llama-index-server/**"
       - ".github/workflows/release_llama_index_server.yml"
   pull_request:
     types:
@@ -19,11 +19,8 @@ jobs:
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: ./python/llama-index-server
-    if: |
-      github.event_name == 'push' && 
-      !startsWith(github.ref, 'refs/heads/release/llama-index-server-v') &&
-      !contains(github.event.head_commit.message, 'Release: llama-index-server v')
+        working-directory: ./llama-index-server
+    if: github.event_name == 'push'
 
     steps:
       - name: Checkout Repository
@@ -31,19 +28,17 @@ jobs:
         with:
           fetch-depth: 0
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
 
+      - name: Install Poetry
+        run: |
+          curl -sSL https://install.python-poetry.org | python3 -
+
       - name: Install dependencies
-        shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install
 
       - name: Setup Git
         run: |
@@ -51,17 +46,15 @@ jobs:
           git config --global user.name "github-actions[bot]"
 
       - name: Bump patch version
-        shell: bash
         run: |
-          uvx --from=toml-cli toml set --toml-path=pyproject.toml project.version $(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version | awk -F. '{$NF = $NF + 1;}1' OFS=.)
+          poetry version patch
           git add pyproject.toml
-          git commit -m "chore(release): bump llama-index-server version to $(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)"
+          git commit -m "chore(release): bump version to $(poetry version -s)"
 
       - name: Get current version
         id: get_version
-        shell: bash
         run: |
-          version=$(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)
+          version=$(poetry version -s)
           echo "current_version=${version}" >> "$GITHUB_OUTPUT"
 
       - name: Create Release PR
@@ -85,7 +78,7 @@ jobs:
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: ./python/llama-index-server
+        working-directory: ./llama-index-server
     if: |
       github.event_name == 'pull_request' && 
       github.event.pull_request.merged == true && 
@@ -96,34 +89,25 @@ jobs:
       - name: Checkout Repository
         uses: actions/checkout@v4
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
 
-      - name: Install dependencies
-        shell: bash
-        run: uv sync --all-extras
-
-      - name: Get current version
-        id: get_version
-        shell: bash
+      - name: Install Poetry
         run: |
-          version=$(uvx --from=toml-cli toml get --toml-path=pyproject.toml project.version)
-          echo "current_version=${version}" >> "$GITHUB_OUTPUT"
+          curl -sSL https://install.python-poetry.org | python3 -
+
+      - name: Install dependencies
+        run: poetry install
 
       - name: Build package
-        shell: bash
-        run: uv build --no-sources
+        run: poetry build
 
       - name: Publish to PyPI
-        shell: bash
-        run: uv publish --token ${{ secrets.PYPI_TOKEN }}
+        env:
+          POETRY_PYPI_TOKEN_PYPI: ${{ secrets.LLAMA_INDEX_PYPI_TOKEN }}
+        run: poetry publish --build
 
       - name: Create GitHub Release
         uses: softprops/action-gh-release@v2

.github/workflows/test_llama_index_server.yml

@@ -4,6 +4,7 @@ on:
   pull_request:
 
 env:
+  POETRY_VERSION: "1.8.3"
   PYTHON_VERSION: "3.9"
 
 jobs:
@@ -12,7 +13,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     defaults:
       run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
     strategy:
       matrix:
         os: [ubuntu-latest, windows-latest]
@@ -20,80 +21,91 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}
 
-      - name: Set up Python ${{ matrix.python-version }}
+      - name: Set up python ${{ matrix.python-version }}
         uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
+          cache: "poetry"
+
+      - name: Configure Poetry
+        run: |
+          poetry config virtualenvs.create true
+          poetry config virtualenvs.in-project true
+          poetry env use python
 
       - name: Install dependencies
         shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install --with dev
 
       - name: Run unit tests
         shell: bash
-        run: uv run pytest tests
+        run: |
+          poetry run pytest tests
 
   type-check:
     name: Type Check
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
     steps:
       - uses: actions/checkout@v4
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}
 
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: ${{ env.PYTHON_VERSION }}
+          cache: "poetry"
+
+      - name: Configure Poetry
+        run: |
+          poetry config virtualenvs.create true
+          poetry config virtualenvs.in-project true
+          poetry env use python
 
       - name: Install dependencies
         shell: bash
-        run: uv sync --all-extras --dev
+        run: poetry install --with dev
 
       - name: Run mypy
         shell: bash
-        run: uv run mypy llama_index
+        run: poetry run mypy llama_index
 
   build:
     needs: [unit-test, type-check]
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: python/llama-index-server
+        working-directory: llama-index-server
     steps:
       - uses: actions/checkout@v4
-
-      - name: Install uv
-        uses: astral-sh/setup-uv@v5
-        with:
-          enable-cache: true
-
+      - name: Install Poetry
+        run: pipx install poetry==${{ env.POETRY_VERSION }}
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: ${{ env.PYTHON_VERSION }}
-
-      - name: Install build package
+      - name: Clear python cache
         shell: bash
-        run: uv sync --all-extras
-
+        run: poetry cache clear --all pypi
+      - name: Build package
+        shell: bash
+        run: poetry build
+      - name: Test installing built package
+        shell: bash
+        run: python -m pip install .
       - name: Test import
         shell: bash
-        run: uv run python -c "from llama_index.server import LlamaIndexServer"
-
+        working-directory: ${{ vars.RUNNER_TEMP }}
+        run: python -c "from llama_index.server import LlamaIndexServer"
       - name: Upload artifact
         uses: actions/upload-artifact@v4
         with:
           name: llama-index-server
-          path: python/llama-index-server/dist/
+          path: llama-index-server/dist/

.gitignore

@@ -6,6 +6,10 @@ node_modules
 .pnpm-store
 .pnp.js
 
+# testing
+coverage
+.coverage
+
 # next.js
 .next/
 out/
@@ -31,9 +35,31 @@ yarn-error.log*
 dist/
 lib/
 
+# e2e
+.cache
+test-results/
+playwright-report/
+blob-report/
+playwright/.cache/
+.tsbuildinfo
+e2e/cache
+
 # intellij
 **/.idea
 
+# Python
+.mypy_cache/
+venv/
+.venv/
+dist/
+.__pycache__
+__pycache__
+.python-version
+.ui
+
+# build artifacts
+create-llama-*.tgz
+
 # vscode
 .vscode
 !.vscode/settings.json

.husky/pre-commit

@@ -1,4 +1,3 @@
 pnpm format
 pnpm lint
-uvx ruff check .
-uvx ruff format . --check 
+uvx ruff format --check templates/

.prettierignore

@@ -1,15 +1,6 @@
-node_modules/
+apps/docs/i18n
+apps/docs/docs/api
 pnpm-lock.yaml
 lib/
 dist/
-cache/
-build/
-.next/
-out/
-packages/server/server/
-
-# Python
-python/
-**/*.mypy_cache/**
-**/*.venv/**
-**/*.ruff_cache/**
+.docusaurus/

CHANGELOG.md

@@ -1,90 +1,5 @@
 # create-llama
 
-## 0.5.12
-
-### Patch Changes
-
-- 241d82a: Add artifacts use case (python)
-
-## 0.5.11
-
-### Patch Changes
-
-- 3960618: chore: create-llama monorepo
-- 8fe5fc2: chore: add llamaindex server package
-
-## 0.5.10
-
-### Patch Changes
-
-- 0a2e12a: Use uv as the default package manager
-
-## 0.5.9
-
-### Patch Changes
-
-- 4bc53ac: Bump new chat ui and update deep research component
-- 4bc53ac: Support generate UI for deep research use case (Typescript)
-
-## 0.5.8
-
-### Patch Changes
-
-- 765181a: chore: test typescript e2e with node 20 and 22
-
-## 0.5.7
-
-### Patch Changes
-
-- 5988657: chore: bump llmaindex
-
-## 0.5.6
-
-### Patch Changes
-
-- d363ced: Bump llamaindex server packages
-
-## 0.5.5
-
-### Patch Changes
-
-- ee85320: The default custom deep research component does not work.
-
-## 0.5.4
-
-### Patch Changes
-
-- 7c3b279: Support code generation of event components using an LLM (Python)
-
-## 0.5.3
-
-### Patch Changes
-
-- 76ec360: Update templates to use new chat ui config
-
-## 0.5.2
-
-### Patch Changes
-
-- c9f8f8d: Use custom component for deep research use case
-
-## 0.5.1
-
-### Patch Changes
-
-- 08b3e07: Simplify the local index code.
-
-## 0.5.0
-
-### Minor Changes
-
-- 54c9e2f: Simplified generated code using LlamaIndexServer
-
-### Patch Changes
-
-- 0e4ecfa: fix: add trycatch for generating error
-- ee69ce7: bump: chat-ui and tailwind v4
-
 ## 0.4.0
 
 ### Minor Changes

README.md

@@ -55,7 +55,7 @@ Then re-start your app. Remember you'll need to re-run `generate` if you add new
 If you're using the Python backend, you can trigger indexing of your data by calling:
 
 ```bash
-uv run generate
+poetry run generate
 ```
 
 ## Customizing the AI models
@@ -130,11 +130,4 @@ Pro mode is ideal for developers who want fine-grained control over their projec
 - [TS/JS docs](https://ts.llamaindex.ai/)
 - [Python docs](https://docs.llamaindex.ai/en/stable/)
 
-## LlamaIndex Server
-
-The generated code is using the LlamaIndex Server, which serves LlamaIndex Workflows and Agent Workflows via an API server. See the following docs for more information:
-
-- [LlamaIndex Server For TypeScript](./packages/server/README.md)
-- [LlamaIndex Server For Python](./python/llama-index-server/README.md)
-
 Inspired by and adapted from [create-next-app](https://github.com/vercel/next.js/tree/canary/packages/create-next-app)

create-app.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import path from "path";
 import { green, yellow } from "picocolors";
 import { tryGitInit } from "./helpers/git";
@@ -89,7 +90,7 @@ export async function createApp({
   // Install backend
   await installTemplate({ ...args, backend: true });
 
-  if (frontend && framework === "fastapi" && template !== "llamaindexserver") {
+  if (frontend && framework === "fastapi") {
     // install frontend
     const frontendRoot = path.join(root, ".frontend");
     await makeDir(frontendRoot);
@@ -109,7 +110,7 @@ export async function createApp({
     console.log();
   }
 
-  if (toolsRequireConfig(tools) && template !== "llamaindexserver") {
+  if (toolsRequireConfig(tools)) {
     const configFile =
       framework === "fastapi" ? "config/tools.yaml" : "config/tools.json";
     console.log(

e2e/python/resolve_dependencies.spec.ts

@@ -195,47 +195,32 @@ async function createAndCheckLlamaProject({
   const pyprojectPath = path.join(projectPath, "pyproject.toml");
   expect(fs.existsSync(pyprojectPath)).toBeTruthy();
 
-  // Modify environment for the command
-  const commandEnv = {
+  const env = {
     ...process.env,
+    POETRY_VIRTUALENVS_IN_PROJECT: "true",
   };
 
-  console.log("Running uv venv...");
+  // Run poetry install
   try {
-    const { stdout: venvStdout, stderr: venvStderr } = await execAsync(
-      "uv venv",
-      { cwd: projectPath, env: commandEnv },
+    const { stdout: installStdout, stderr: installStderr } = await execAsync(
+      "poetry install",
+      { cwd: projectPath, env },
     );
-    console.log("uv venv stdout:", venvStdout);
-    console.error("uv venv stderr:", venvStderr);
+    console.log("poetry install stdout:", installStdout);
+    console.error("poetry install stderr:", installStderr);
   } catch (error) {
-    console.error("Error running uv venv:", error);
-    throw error; // Re-throw error to fail the test
+    console.error("Error running poetry install:", error);
+    throw error;
   }
 
-  console.log("Running uv sync...");
-  try {
-    const { stdout: syncStdout, stderr: syncStderr } = await execAsync(
-      "uv sync --all-extras",
-      { cwd: projectPath, env: commandEnv },
-    );
-    console.log("uv sync stdout:", syncStdout);
-    console.error("uv sync stderr:", syncStderr);
-  } catch (error) {
-    console.error("Error running uv sync:", error);
-    throw error; // Re-throw error to fail the test
-  }
-
-  console.log("Running uv run mypy ....");
+  // Run poetry run mypy
   try {
     const { stdout: mypyStdout, stderr: mypyStderr } = await execAsync(
-      "uv run mypy .",
-      { cwd: projectPath, env: commandEnv },
+      "poetry run mypy .",
+      { cwd: projectPath, env },
     );
-    console.log("uv run mypy stdout:", mypyStdout);
-    console.error("uv run mypy stderr:", mypyStderr);
-    // Assuming mypy success means no output or specific success message
-    // Adjust checks based on actual expected mypy output
+    console.log("poetry run mypy stdout:", mypyStdout);
+    console.error("poetry run mypy stderr:", mypyStderr);
   } catch (error) {
     console.error("Error running mypy:", error);
     throw error;

e2e/shared/multiagent_template.spec.ts

@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";
@@ -15,17 +16,15 @@ const templateFramework: TemplateFramework = process.env.FRAMEWORK
 const dataSource: string = "--example-file";
 const templateUI: TemplateUI = "shadcn";
 const templatePostInstallAction: TemplatePostInstallAction = "runApp";
-const appType: AppType = "--frontend";
+const appType: AppType = templateFramework === "fastapi" ? "--frontend" : "";
 const userMessage = "Write a blog post about physical standards for letters";
-const templateUseCases = ["financial_report", "agentic_rag", "deep_research"];
+const templateUseCases = ["financial_report", "blog", "form_filling"];
 
 for (const useCase of templateUseCases) {
-  test.describe(`Test use case ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
+  test.describe(`Test multiagent template ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
     test.skip(
-      process.platform !== "linux" ||
-        process.env.DATASOURCE === "--no-files" ||
-        templateFramework === "express",
-      "The llamaindexserver template currently only works with nextjs, fastapi. We also only run on Linux to speed up tests.",
+      process.platform !== "linux" || process.env.DATASOURCE === "--no-files",
+      "The multiagent template currently only works with files. We also only run on Linux to speed up tests.",
     );
     let port: number;
     let cwd: string;
@@ -39,7 +38,7 @@ for (const useCase of templateUseCases) {
       cwd = await createTestDir();
       const result = await runCreateLlama({
         cwd,
-        templateType: "llamaindexserver",
+        templateType: "multiagent",
         templateFramework,
         dataSource,
         vectorDb,
@@ -64,9 +63,7 @@ for (const useCase of templateUseCases) {
           templateFramework === "express",
       );
       await page.goto(`http://localhost:${port}`);
-      await expect(page.getByText("Built by LlamaIndex")).toBeVisible({
-        timeout: 5 * 60 * 1000,
-      });
+      await expect(page.getByText("Built by LlamaIndex")).toBeVisible();
     });
 
     test("Frontend should be able to submit a message and receive the start of a streamed response", async ({
@@ -75,9 +72,9 @@ for (const useCase of templateUseCases) {
       test.skip(
         templatePostInstallAction !== "runApp" ||
           useCase === "financial_report" ||
-          useCase === "deep_research" ||
+          useCase === "form_filling" ||
           templateFramework === "express",
-        "Skip chat tests for financial report and deep research.",
+        "Skip chat tests for financial report and form filling.",
       );
       await page.goto(`http://localhost:${port}`);
       await page.fill("form textarea", userMessage);
@@ -89,12 +86,6 @@ for (const useCase of templateUseCases) {
       await page.click("form button[type=submit]");
 
       const response = await responsePromise;
-      console.log(`Response status: ${response.status()}`);
-      const responseBody = await response
-        .text()
-        .catch((e) => `Error reading body: ${e}`);
-      console.log(`Response body: ${responseBody}`);
-
       expect(response.ok()).toBeTruthy();
     });
 

e2e/shared/reflex_template.spec.ts

@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";

e2e/shared/streaming_template.spec.ts

@@ -1,3 +1,4 @@
+/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";

e2e/typescript/resolve_dependencies.spec.ts

@@ -74,7 +74,7 @@ test.describe("Test resolve TS dependencies", () => {
     // Install dependencies using pnpm
     try {
       const { stderr: installStderr } = await execAsync(
-        "pnpm install --prefer-offline --ignore-workspace",
+        "pnpm install --prefer-offline",
         {
           cwd: appDir,
         },

e2e/utils.ts

@@ -67,8 +67,8 @@ export async function runCreateLlama({
   ].join("-");
 
   // Handle different data source types
-  const dataSourceArgs = [];
-  if (dataSource.includes("--web-source")) {
+  let dataSourceArgs = [];
+  if (dataSource.includes("--web-source" || "--db-source")) {
     const webSource = dataSource.split(" ")[1];
     dataSourceArgs.push("--web-source", webSource);
   } else if (dataSource.includes("--db-source")) {
@@ -113,12 +113,7 @@ export async function runCreateLlama({
   if (observability) {
     commandArgs.push("--observability", observability);
   }
-  if (
-    (templateType === "multiagent" ||
-      templateType === "reflex" ||
-      templateType === "llamaindexserver") &&
-    useCase
-  ) {
+  if ((templateType === "multiagent" || templateType === "reflex") && useCase) {
     commandArgs.push("--use-case", useCase);
   }
 

eslint.config.mjs

@@ -1,62 +0,0 @@
-import eslint from "@eslint/js";
-import eslintConfigPrettier from "eslint-config-prettier";
-import globals from "globals";
-import tseslint from "typescript-eslint";
-
-export default tseslint.config(
-  eslint.configs.recommended,
-  ...tseslint.configs.recommended,
-  eslintConfigPrettier,
-  {
-    languageOptions: {
-      ecmaVersion: 2022,
-      sourceType: "module",
-      globals: {
-        ...globals.browser,
-        ...globals.node,
-      },
-    },
-  },
-  {
-    files: ["packages/create-llama/**"],
-    rules: {
-      "max-params": ["error", 4],
-      "prefer-const": "error",
-      "no-empty": "off",
-      "no-extra-boolean-cast": "off",
-      "@typescript-eslint/no-explicit-any": "off",
-      "@typescript-eslint/no-unused-vars": "off",
-      "@typescript-eslint/no-empty-object-type": "off",
-      "@typescript-eslint/no-wrapper-object-types": "off",
-      "@typescript-eslint/ban-ts-comment": "off",
-    },
-  },
-  {
-    files: ["packages/server/**"],
-    rules: {
-      "no-irregular-whitespace": "off",
-      "@typescript-eslint/no-unused-vars": "off",
-      "@typescript-eslint/no-explicit-any": [
-        "error",
-        {
-          ignoreRestArgs: true,
-        },
-      ],
-    },
-  },
-  {
-    ignores: [
-      "python/**",
-      "**/*.mypy_cache/**",
-      "**/*.venv/**",
-      "**/*.ruff_cache/**",
-      "**/dist/**",
-      "**/e2e/cache/**",
-      "**/lib/*",
-      "**/.next/**",
-      "**/out/**",
-      "**/node_modules/**",
-      "**/build/**",
-    ],
-  },
-);

helpers/copy.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { async as glob } from "fast-glob";
 import fs from "fs";
 import path from "path";

helpers/env-variables.ts

@@ -44,7 +44,6 @@ const renderEnvVar = (envVars: EnvVar[]): string => {
 const getVectorDBEnvs = (
   vectorDb?: TemplateVectorDB,
   framework?: TemplateFramework,
-  template?: TemplateType,
 ): EnvVar[] => {
   if (!vectorDb || !framework) {
     return [];
@@ -169,7 +168,7 @@ const getVectorDBEnvs = (
           description:
             "The organization ID for the LlamaCloud project (uses default organization if not specified)",
         },
-        ...(framework === "nextjs" && template !== "llamaindexserver"
+        ...(framework === "nextjs"
           ? // activate index selector per default (not needed for non-NextJS backends as it's handled by createFrontendEnvFile)
             [
               {
@@ -181,7 +180,7 @@ const getVectorDBEnvs = (
             ]
           : []),
       ];
-    case "chroma": {
+    case "chroma":
       const envs = [
         {
           name: "CHROMA_COLLECTION",
@@ -206,7 +205,6 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
         });
       }
       return envs;
-    }
     case "weaviate":
       return [
         {
@@ -225,15 +223,13 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
         },
       ];
     default:
-      return template !== "llamaindexserver"
-        ? [
-            {
-              name: "STORAGE_CACHE_DIR",
-              description: "The directory to store the local storage cache.",
-              value: ".cache",
-            },
-          ]
-        : [];
+      return [
+        {
+          name: "STORAGE_CACHE_DIR",
+          description: "The directory to store the local storage cache.",
+          value: ".cache",
+        },
+      ];
   }
 };
 
@@ -386,42 +382,38 @@ const getModelEnvs = (modelConfig: ModelConfig): EnvVar[] => {
 
 const getFrameworkEnvs = (
   framework: TemplateFramework,
-  template: TemplateType,
   port?: number,
 ): EnvVar[] => {
   const sPort = port?.toString() || "8000";
-  const result: EnvVar[] =
-    template !== "llamaindexserver"
-      ? [
-          {
-            name: "FILESERVER_URL_PREFIX",
-            description:
-              "FILESERVER_URL_PREFIX is the URL prefix of the server storing the images generated by the interpreter.",
-            value:
-              framework === "nextjs"
-                ? // FIXME: if we are using nextjs, port should be 3000
-                  "http://localhost:3000/api/files"
-                : `http://localhost:${sPort}/api/files`,
-          },
-        ]
-      : [];
+  const result: EnvVar[] = [
+    {
+      name: "FILESERVER_URL_PREFIX",
+      description:
+        "FILESERVER_URL_PREFIX is the URL prefix of the server storing the images generated by the interpreter.",
+      value:
+        framework === "nextjs"
+          ? // FIXME: if we are using nextjs, port should be 3000
+            "http://localhost:3000/api/files"
+          : `http://localhost:${sPort}/api/files`,
+    },
+  ];
   if (framework === "fastapi") {
     result.push(
       ...[
         {
           name: "APP_HOST",
-          description: "The address to start the FastAPI app.",
+          description: "The address to start the backend app.",
           value: "0.0.0.0",
         },
         {
           name: "APP_PORT",
-          description: "The port to start the FastAPI app.",
+          description: "The port to start the backend app.",
           value: sPort,
         },
       ],
     );
   }
-  if (framework === "nextjs" && template !== "llamaindexserver") {
+  if (framework === "nextjs") {
     result.push({
       name: "NEXT_PUBLIC_CHAT_API",
       description:
@@ -577,41 +569,25 @@ export const createBackendEnvFile = async (
     | "port"
     | "tools"
     | "observability"
-    | "useLlamaParse"
   >,
 ) => {
   // Init env values
   const envFileName = ".env";
   const envVars: EnvVar[] = [
-    ...(opts.useLlamaParse
-      ? [
-          {
-            name: "LLAMA_CLOUD_API_KEY",
-            description: `The Llama Cloud API key.`,
-            value: opts.llamaCloudKey,
-          },
-        ]
-      : []),
-    ...getVectorDBEnvs(opts.vectorDb, opts.framework, opts.template),
-    ...getToolEnvs(opts.tools),
-    ...getFrameworkEnvs(opts.framework, opts.template, opts.port),
+    {
+      name: "LLAMA_CLOUD_API_KEY",
+      description: `The Llama Cloud API key.`,
+      value: opts.llamaCloudKey,
+    },
     // Add environment variables of each component
-    ...(opts.template === "llamaindexserver"
-      ? [
-          {
-            name: "OPENAI_API_KEY",
-            description: "The OpenAI API key to use.",
-            value: opts.modelConfig.apiKey,
-          },
-        ]
-      : [
-          // don't use this stuff for llama-indexserver
-          ...getModelEnvs(opts.modelConfig),
-          ...getEngineEnvs(),
-          ...getTemplateEnvs(opts.template),
-          ...getObservabilityEnvs(opts.observability),
-          ...getSystemPromptEnv(opts.tools, opts.dataSources, opts.template),
-        ]),
+    ...getModelEnvs(opts.modelConfig),
+    ...getEngineEnvs(),
+    ...getVectorDBEnvs(opts.vectorDb, opts.framework),
+    ...getFrameworkEnvs(opts.framework, opts.port),
+    ...getToolEnvs(opts.tools),
+    ...getTemplateEnvs(opts.template),
+    ...getObservabilityEnvs(opts.observability),
+    ...getSystemPromptEnv(opts.tools, opts.dataSources, opts.template),
   ];
   // Render and write env file
   const content = renderEnvVar(envVars);

helpers/git.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import fs from "fs";
 import path from "path";

helpers/index.ts

@@ -1,7 +1,7 @@
 import { callPackageManager } from "./install";
 
 import path from "path";
-import picocolors, { cyan } from "picocolors";
+import { cyan } from "picocolors";
 
 import fsExtra from "fs-extra";
 import { writeLoadersConfig } from "./datasources";
@@ -9,6 +9,7 @@ import { createBackendEnvFile, createFrontendEnvFile } from "./env-variables";
 import { PackageManager } from "./get-pkg-manager";
 import { installLlamapackProject } from "./llama-pack";
 import { makeDir } from "./make-dir";
+import { isHavingPoetryLockFile, tryPoetryRun } from "./poetry";
 import { installPythonTemplate } from "./python";
 import { downloadAndExtractRepo } from "./repo";
 import { ConfigFileType, writeToolsConfig } from "./tools";
@@ -21,7 +22,6 @@ import {
   TemplateVectorDB,
 } from "./types";
 import { installTSTemplate } from "./typescript";
-import { isHavingUvLockFile, tryUvRun } from "./uv";
 
 const checkForGenerateScript = (
   modelConfig: ModelConfig,
@@ -41,11 +41,7 @@ const checkForGenerateScript = (
     missingSettings.push("your LLAMA_CLOUD_API_KEY");
   }
 
-  if (
-    vectorDb !== undefined &&
-    vectorDb !== "none" &&
-    vectorDb !== "llamacloud"
-  ) {
+  if (vectorDb !== "none" && vectorDb !== "llamacloud") {
     missingSettings.push("your Vector DB environment variables");
   }
 
@@ -64,7 +60,7 @@ async function generateContextData(
   if (packageManager) {
     const runGenerate = `${cyan(
       framework === "fastapi"
-        ? "uv run generate"
+        ? "poetry run generate"
         : `${packageManager} run generate`,
     )}`;
 
@@ -78,21 +74,15 @@ async function generateContextData(
     if (!missingSettings.length) {
       // If all the required environment variables are set, run the generate script
       if (framework === "fastapi") {
-        if (isHavingUvLockFile()) {
+        if (isHavingPoetryLockFile()) {
           console.log(`Running ${runGenerate} to generate the context data.`);
-          const result = tryUvRun("generate");
+          const result = tryPoetryRun("poetry run generate");
           if (!result) {
             console.log(`Failed to run ${runGenerate}.`);
             process.exit(1);
           }
           console.log(`Generated context data`);
           return;
-        } else {
-          console.log(
-            picocolors.yellow(
-              `\nWarning: uv.lock not found. Dependency installation might be incomplete. Skipping context generation.\nIf dependencies were installed, try running '${runGenerate}' manually.\n`,
-            ),
-          );
         }
       } else {
         console.log(`Running ${runGenerate} to generate the context data.`);
@@ -102,14 +92,14 @@ async function generateContextData(
     }
 
     const settingsMessage = `After setting ${missingSettings.join(" and ")}, run ${runGenerate} to generate the context data.`;
-    console.log(picocolors.yellow(`\n${settingsMessage}\n\n`));
+    console.log(`\n${settingsMessage}\n\n`);
   }
 }
 
 const downloadFile = async (url: string, destPath: string) => {
   const response = await fetch(url);
   const fileBuffer = await response.arrayBuffer();
-  await fsExtra.writeFile(destPath, new Uint8Array(fileBuffer));
+  await fsExtra.writeFile(destPath, Buffer.from(fileBuffer));
 };
 
 const prepareContextData = async (
@@ -176,17 +166,6 @@ export const installTemplate = async (
 
   if (props.framework === "fastapi") {
     await installPythonTemplate(props);
-  } else {
-    await installTSTemplate(props);
-  }
-
-  // write configurations
-  if (props.template !== "llamaindexserver") {
-    await writeToolsConfig(
-      props.root,
-      props.tools,
-      props.framework === "fastapi" ? ConfigFileType.YAML : ConfigFileType.JSON,
-    );
     if (props.vectorDb !== "llamacloud") {
       // write loaders configuration (currently Python only)
       // not needed for LlamaCloud as it has its own loaders
@@ -196,13 +175,26 @@ export const installTemplate = async (
         props.useLlamaParse,
       );
     }
+  } else {
+    await installTSTemplate(props);
   }
 
+  // write tools configuration
+  await writeToolsConfig(
+    props.root,
+    props.tools,
+    props.framework === "fastapi" ? ConfigFileType.YAML : ConfigFileType.JSON,
+  );
+
   if (props.backend) {
     // This is a backend, so we need to copy the test data and create the env file.
 
     // Copy the environment file to the target directory.
-    if (props.template !== "community" && props.template !== "llamapack") {
+    if (
+      props.template === "streaming" ||
+      props.template === "multiagent" ||
+      props.template === "reflex"
+    ) {
       await createBackendEnvFile(props.root, props);
     }
 

helpers/install.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import spawn from "cross-spawn";
 import { yellow } from "picocolors";
 import type { PackageManager } from "./get-pkg-manager";

helpers/is-folder-empty.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import fs from "fs";
 import path from "path";
 import { blue, green } from "picocolors";

helpers/llama-pack.ts

@@ -143,6 +143,6 @@ export const installLlamapackProject = async ({
   await copyData({ root });
   await installLlamapackExample({ root, llamapack });
   if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
-    installPythonDependencies();
+    installPythonDependencies({ noRoot: true });
   }
 };

helpers/poetry.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import fs from "fs";
 

helpers/providers/index.ts

@@ -28,7 +28,7 @@ export async function askModelConfig({
 }: ModelConfigQuestionsParams): Promise<ModelConfig> {
   let modelProvider: ModelProvider = DEFAULT_MODEL_PROVIDER;
   if (askModels) {
-    const choices = [
+    let choices = [
       { title: "OpenAI", value: "openai" },
       { title: "Groq", value: "groq" },
       { title: "Ollama", value: "ollama" },

helpers/python.ts

@@ -3,16 +3,15 @@ import path from "path";
 import { cyan, red } from "picocolors";
 import { parse, stringify } from "smol-toml";
 import terminalLink from "terminal-link";
-import { isUvAvailable, tryUvSync } from "./uv";
 
 import { assetRelocator, copy } from "./copy";
 import { templatesDir } from "./dir";
+import { isPoetryAvailable, tryPoetryInstall } from "./poetry";
 import { Tool } from "./tools";
 import {
   InstallTemplateArgs,
   ModelConfig,
   TemplateDataSource,
-  TemplateObservability,
   TemplateType,
   TemplateVectorDB,
 } from "./types";
@@ -30,8 +29,6 @@ const getAdditionalDependencies = (
   dataSources?: TemplateDataSource[],
   tools?: Tool[],
   templateType?: TemplateType,
-  observability?: TemplateObservability,
-  // eslint-disable-next-line max-params
 ) => {
   const dependencies: Dependency[] = [];
 
@@ -40,21 +37,21 @@ const getAdditionalDependencies = (
     case "mongo": {
       dependencies.push({
         name: "llama-index-vector-stores-mongodb",
-        version: ">=0.3.2,<0.4.0",
+        version: "^0.6.0",
       });
       break;
     }
     case "pg": {
       dependencies.push({
         name: "llama-index-vector-stores-postgres",
-        version: ">=0.3.2,<0.4.0",
+        version: "^0.3.2",
       });
       break;
     }
     case "pinecone": {
       dependencies.push({
         name: "llama-index-vector-stores-pinecone",
-        version: ">=0.4.1,<0.5.0",
+        version: "^0.4.1",
         constraints: {
           python: ">=3.11,<3.13",
         },
@@ -64,25 +61,25 @@ const getAdditionalDependencies = (
     case "milvus": {
       dependencies.push({
         name: "llama-index-vector-stores-milvus",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
       });
       dependencies.push({
         name: "pymilvus",
-        version: ">=2.4.4,<3.0.0",
+        version: "2.4.4",
       });
       break;
     }
     case "astra": {
       dependencies.push({
         name: "llama-index-vector-stores-astra-db",
-        version: ">=0.4.0,<0.5.0",
+        version: "^0.4.0",
       });
       break;
     }
     case "qdrant": {
       dependencies.push({
         name: "llama-index-vector-stores-qdrant",
-        version: ">=0.4.0,<0.5.0",
+        version: "^0.4.0",
         constraints: {
           python: ">=3.11,<3.13",
         },
@@ -92,21 +89,21 @@ const getAdditionalDependencies = (
     case "chroma": {
       dependencies.push({
         name: "llama-index-vector-stores-chroma",
-        version: ">=0.4.0,<0.5.0",
+        version: "^0.4.0",
       });
       break;
     }
     case "weaviate": {
       dependencies.push({
         name: "llama-index-vector-stores-weaviate",
-        version: ">=1.2.3,<2.0.0",
+        version: "^1.2.3",
       });
       break;
     }
     case "llamacloud":
       dependencies.push({
         name: "llama-index-indices-managed-llama-cloud",
-        version: ">=0.6.3,<0.7.0",
+        version: "^0.6.3",
       });
       break;
   }
@@ -119,28 +116,28 @@ const getAdditionalDependencies = (
         case "file":
           dependencies.push({
             name: "docx2txt",
-            version: ">=0.8,<0.9",
+            version: "^0.8",
           });
           break;
         case "web":
           dependencies.push({
             name: "llama-index-readers-web",
-            version: ">=0.3.0,<0.4.0",
+            version: "^0.3.0",
           });
           break;
         case "db":
           dependencies.push({
             name: "llama-index-readers-database",
-            version: ">=0.3.0,<0.4.0",
+            version: "^0.3.0",
           });
           dependencies.push({
             name: "pymysql",
-            version: ">=1.1.0,<2.0.0",
+            version: "^1.1.0",
             extras: ["rsa"],
           });
           dependencies.push({
             name: "psycopg2-binary",
-            version: ">=2.9.9,<3.0.0",
+            version: "^2.9.9",
           });
           break;
       }
@@ -159,122 +156,155 @@ const getAdditionalDependencies = (
     case "ollama":
       dependencies.push({
         name: "llama-index-llms-ollama",
-        version: ">=0.5.0,<0.6.0",
+        version: "0.3.0",
       });
       dependencies.push({
         name: "llama-index-embeddings-ollama",
-        version: ">=0.6.0,<0.7.0",
+        version: "0.3.0",
       });
       break;
     case "openai":
       if (templateType !== "multiagent") {
         dependencies.push({
           name: "llama-index-llms-openai",
-          version: ">=0.3.2,<0.4.0",
+          version: "^0.3.2",
         });
         dependencies.push({
           name: "llama-index-embeddings-openai",
-          version: ">=0.3.1,<0.4.0",
+          version: "^0.3.1",
         });
         dependencies.push({
           name: "llama-index-agent-openai",
-          version: ">=0.4.0,<0.5.0",
+          version: "^0.4.0",
         });
       }
       break;
     case "groq":
+      // Fastembed==0.2.0 does not support python3.13 at the moment
+      // Fixed the python version less than 3.13
+      dependencies.push({
+        name: "python",
+        version: "^3.11,<3.13",
+      });
       dependencies.push({
         name: "llama-index-llms-groq",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
       });
       dependencies.push({
         name: "llama-index-embeddings-fastembed",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
       });
       break;
     case "anthropic":
+      // Fastembed==0.2.0 does not support python3.13 at the moment
+      // Fixed the python version less than 3.13
+      dependencies.push({
+        name: "python",
+        version: "^3.11,<3.13",
+      });
       dependencies.push({
         name: "llama-index-llms-anthropic",
-        version: ">=0.6.0,<0.7.0",
+        version: "0.3.0",
       });
       dependencies.push({
         name: "llama-index-embeddings-fastembed",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
       });
       break;
     case "gemini":
       dependencies.push({
         name: "llama-index-llms-gemini",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.3.4",
       });
       dependencies.push({
         name: "llama-index-embeddings-gemini",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.2.0",
       });
       break;
     case "mistral":
       dependencies.push({
         name: "llama-index-llms-mistralai",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.2.1",
       });
       dependencies.push({
         name: "llama-index-embeddings-mistralai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
       });
       break;
     case "azure-openai":
       dependencies.push({
         name: "llama-index-llms-azure-openai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
       });
       dependencies.push({
         name: "llama-index-embeddings-azure-openai",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.4",
       });
       break;
     case "huggingface":
       dependencies.push({
         name: "llama-index-llms-huggingface",
-        version: ">=0.5.0,<0.6.0",
+        version: "^0.3.5",
       });
       dependencies.push({
         name: "llama-index-embeddings-huggingface",
-        version: ">=0.5.0,<0.6.0",
+        version: "^0.3.1",
       });
       dependencies.push({
         name: "optimum",
-        version: ">=1.23.3,<2.0.0",
+        version: "^1.23.3",
         extras: ["onnxruntime"],
       });
       break;
     case "t-systems":
       dependencies.push({
         name: "llama-index-agent-openai",
-        version: ">=0.4.0,<0.5.0",
+        version: "0.3.0",
       });
       dependencies.push({
         name: "llama-index-llms-openai-like",
-        version: ">=0.3.0,<0.4.0",
+        version: "0.2.0",
       });
       break;
   }
 
-  if (observability && observability !== "none") {
-    if (observability === "traceloop") {
-      dependencies.push({
-        name: "traceloop-sdk",
-        version: ">=0.15.11,<0.16.0",
-      });
+  return dependencies;
+};
+
+const mergePoetryDependencies = (
+  dependencies: Dependency[],
+  existingDependencies: Record<string, Omit<Dependency, "name"> | string>,
+) => {
+  for (const dependency of dependencies) {
+    let value = existingDependencies[dependency.name] ?? {};
+
+    // default string value is equal to attribute "version"
+    if (typeof value === "string") {
+      value = { version: value };
     }
-    if (observability === "llamatrace") {
-      dependencies.push({
-        name: "llama-index-callbacks-arize-phoenix",
-        version: ">=0.3.0,<0.4.0",
-      });
+
+    value.version = dependency.version ?? value.version;
+    value.extras = dependency.extras ?? value.extras;
+
+    // Merge constraints if they exist
+    if (dependency.constraints) {
+      value = { ...value, ...dependency.constraints };
+    }
+
+    if (value.version === undefined) {
+      throw new Error(
+        `Dependency "${dependency.name}" is missing attribute "version"!`,
+      );
+    }
+
+    // Serialize as object if there are any additional properties
+    if (Object.keys(value).length > 1) {
+      existingDependencies[dependency.name] = value;
+    } else {
+      // Otherwise, serialize just the version string
+      existingDependencies[dependency.name] = value.version;
     }
   }
-
-  return dependencies;
 };
 
 const copyRouterCode = async (root: string, tools: Tool[]) => {
@@ -299,100 +329,19 @@ export const addDependencies = async (
     // Parse toml file
     const file = path.join(projectDir, FILENAME);
     const fileContent = await fs.readFile(file, "utf8");
-    let fileParsed: any;
-    try {
-      fileParsed = parse(fileContent);
-    } catch (parseError) {
-      console.error(`Error parsing ${FILENAME}:`, parseError);
-      throw new Error(
-        `Failed to parse ${FILENAME}. Please ensure it's valid TOML.`,
-      );
-    }
+    const fileParsed = parse(fileContent);
 
-    // Ensure [project] and [project.dependencies] exist
-    if (!fileParsed.project) {
-      fileParsed.project = {};
-    }
-    if (
-      !fileParsed.project.dependencies ||
-      !Array.isArray(fileParsed.project.dependencies)
-    ) {
-      // If dependencies exist but aren't an array, log a warning or error.
-      // For now, we'll overwrite it, assuming the intent is to use the standard array format.
-      console.warn(
-        `[project.dependencies] in ${FILENAME} is not an array. It will be overwritten.`,
-      );
-      fileParsed.project.dependencies = [];
-    }
-
-    const existingDependencies: string[] = fileParsed.project.dependencies;
-    const addedDeps: string[] = [];
-    const updatedDeps: string[] = [];
-
-    // Add or update dependencies
-    for (const newDep of dependencies) {
-      let depString = newDep.name;
-      if (newDep.extras && newDep.extras.length > 0) {
-        depString += `[${newDep.extras.join(",")}]`;
-      }
-      if (newDep.version) {
-        depString += newDep.version;
-      }
-
-      let found = false;
-      for (let i = 0; i < existingDependencies.length; i++) {
-        const existingDepNameMatch =
-          existingDependencies[i].match(/^([a-zA-Z0-9._-]+)/);
-        if (
-          existingDepNameMatch &&
-          existingDepNameMatch[1].toLowerCase() === depString.toLowerCase()
-        ) {
-          // Found existing dependency, update it
-          if (existingDependencies[i] !== depString) {
-            updatedDeps.push(`${existingDependencies[i]} -> ${depString}`);
-            existingDependencies[i] = depString;
-          }
-          found = true;
-          break;
-        }
-      }
-
-      if (!found) {
-        // Add new dependency
-        existingDependencies.push(depString);
-        addedDeps.push(depString);
-      }
-      // Handle python version constraints separately (if any)
-      if (newDep.constraints?.python) {
-        if (
-          !fileParsed.project["requires-python"] ||
-          fileParsed.project["requires-python"] !== newDep.constraints.python
-        ) {
-          // This simple overwrite might not be ideal; merging constraints is complex.
-          // For now, let's just set it if the new dependency has one.
-          console.log(
-            `Setting requires-python = "${newDep.constraints.python}" from dependency ${newDep.name}`,
-          );
-          fileParsed.project["requires-python"] = newDep.constraints.python;
-        }
-      }
-    }
+    // Modify toml dependencies
+    const tool = fileParsed.tool as any;
+    const existingDependencies = tool.poetry.dependencies;
+    mergePoetryDependencies(dependencies, existingDependencies);
 
     // Write toml file
     const newFileContent = stringify(fileParsed);
     await fs.writeFile(file, newFileContent);
 
-    if (addedDeps.length > 0) {
-      console.log(`\nAdded dependencies to ${cyan(FILENAME)}:`);
-      addedDeps.forEach((dep) => console.log(`  ${dep}`));
-    }
-    if (updatedDeps.length > 0) {
-      console.log(`\nUpdated dependencies in ${cyan(FILENAME)}:`);
-      updatedDeps.forEach((dep) => console.log(`  ${dep}`));
-    }
-    if (addedDeps.length > 0 || updatedDeps.length > 0) {
-      console.log(""); // Newline for spacing
-    }
+    const dependenciesString = dependencies.map((d) => d.name).join(", ");
+    console.log(`\nAdded ${dependenciesString} to ${cyan(FILENAME)}\n`);
   } catch (error) {
     console.log(
       `Error while updating dependencies for Poetry project file ${FILENAME}\n`,
@@ -401,16 +350,18 @@ export const addDependencies = async (
   }
 };
 
-export const installPythonDependencies = () => {
-  if (isUvAvailable()) {
+export const installPythonDependencies = (
+  { noRoot }: { noRoot: boolean } = { noRoot: false },
+) => {
+  if (isPoetryAvailable()) {
     console.log(
-      `Installing Python dependencies using uv. This may take a while...`,
+      `Installing python dependencies using poetry. This may take a while...`,
     );
-    const installSuccessful = tryUvSync();
+    const installSuccessful = tryPoetryInstall(noRoot);
     if (!installSuccessful) {
       console.error(
         red(
-          "Installing dependencies using uv failed. Please check the error log above and ensure uv is installed correctly.",
+          "Installing dependencies using poetry failed. Please check error log above and try running create-llama again.",
         ),
       );
       process.exit(1);
@@ -418,34 +369,57 @@ export const installPythonDependencies = () => {
   } else {
     console.error(
       red(
-        `uv is not available in the current environment. Please check ${terminalLink(
-          "uv Installation",
-          `https://github.com/astral-sh/uv#installation`,
-        )} to install uv first, then run create-llama again.`,
+        `Poetry is not available in the current environment. Please check ${terminalLink(
+          "Poetry Installation",
+          `https://python-poetry.org/docs/#installation`,
+        )} to install poetry first, then run create-llama again.`,
       ),
     );
     process.exit(1);
   }
 };
 
-const installLegacyPythonTemplate = async ({
+export const installPythonTemplate = async ({
+  appName,
   root,
   template,
+  framework,
   vectorDb,
+  postInstallAction,
+  modelConfig,
   dataSources,
   tools,
+  useLlamaParse,
   useCase,
   observability,
 }: Pick<
   InstallTemplateArgs,
+  | "appName"
   | "root"
   | "template"
+  | "framework"
   | "vectorDb"
+  | "postInstallAction"
+  | "modelConfig"
   | "dataSources"
   | "tools"
+  | "useLlamaParse"
   | "useCase"
   | "observability"
 >) => {
+  console.log("\nInitializing Python project with template:", template, "\n");
+  let templatePath;
+  if (template === "reflex") {
+    templatePath = path.join(templatesDir, "types", "reflex");
+  } else {
+    templatePath = path.join(templatesDir, "types", "streaming", framework);
+  }
+  await copy("**", root, {
+    parents: true,
+    cwd: templatePath,
+    rename: assetRelocator,
+  });
+
   const compPath = path.join(templatesDir, "components");
   const enginePath = path.join(root, "app", "engine");
 
@@ -535,7 +509,34 @@ const installLegacyPythonTemplate = async ({
     }
   }
 
+  console.log("Adding additional dependencies");
+
+  const addOnDependencies = getAdditionalDependencies(
+    modelConfig,
+    vectorDb,
+    dataSources,
+    tools,
+    template,
+  );
+
   if (observability && observability !== "none") {
+    if (observability === "traceloop") {
+      addOnDependencies.push({
+        name: "traceloop-sdk",
+        version: "^0.15.11",
+      });
+    }
+
+    if (observability === "llamatrace") {
+      addOnDependencies.push({
+        name: "llama-index-callbacks-arize-phoenix",
+        version: "^0.3.0",
+        constraints: {
+          python: ">=3.11,<3.13",
+        },
+      });
+    }
+
     const templateObservabilityPath = path.join(
       templatesDir,
       "components",
@@ -547,133 +548,6 @@ const installLegacyPythonTemplate = async ({
       cwd: templateObservabilityPath,
     });
   }
-};
-
-const installLlamaIndexServerTemplate = async ({
-  root,
-  useCase,
-  useLlamaParse,
-}: Pick<InstallTemplateArgs, "root" | "useCase" | "useLlamaParse">) => {
-  if (!useCase) {
-    console.log(
-      red(
-        `There is no use case selected. Please pick a use case to use via --use-case flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  await copy("*.py", path.join(root, "app"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "workflows", "python", useCase),
-  });
-
-  // Copy custom UI component code
-  await copy(`*`, path.join(root, "components"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "ui", "workflows", useCase),
-  });
-
-  if (useLlamaParse) {
-    await copy("index.py", path.join(root, "app"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "python",
-      ),
-    });
-    // TODO: Consider moving generate.py to app folder.
-    await copy("generate.py", path.join(root), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "python",
-      ),
-    });
-  }
-  // Copy README.md
-  await copy("README-template.md", path.join(root), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "workflows", "python", useCase),
-    rename: assetRelocator,
-  });
-};
-
-export const installPythonTemplate = async ({
-  appName,
-  root,
-  template,
-  framework,
-  vectorDb,
-  postInstallAction,
-  modelConfig,
-  dataSources,
-  tools,
-  useLlamaParse,
-  useCase,
-  observability,
-}: Pick<
-  InstallTemplateArgs,
-  | "appName"
-  | "root"
-  | "template"
-  | "framework"
-  | "vectorDb"
-  | "postInstallAction"
-  | "modelConfig"
-  | "dataSources"
-  | "tools"
-  | "useLlamaParse"
-  | "useCase"
-  | "observability"
->) => {
-  console.log("\nInitializing Python project with template:", template, "\n");
-  let templatePath;
-  if (template === "reflex") {
-    templatePath = path.join(templatesDir, "types", "reflex");
-  } else {
-    templatePath = path.join(templatesDir, "types", template, framework);
-  }
-  await copy("**", root, {
-    parents: true,
-    cwd: templatePath,
-    rename: assetRelocator,
-  });
-
-  if (template === "llamaindexserver") {
-    await installLlamaIndexServerTemplate({
-      root,
-      useCase,
-      useLlamaParse,
-    });
-  } else {
-    await installLegacyPythonTemplate({
-      root,
-      template,
-      vectorDb,
-      dataSources,
-      tools,
-      useCase,
-      observability,
-    });
-  }
-
-  console.log("Adding additional dependencies");
-  const addOnDependencies = getAdditionalDependencies(
-    modelConfig,
-    vectorDb,
-    dataSources,
-    tools,
-    template,
-  );
 
   await addDependencies(root, addOnDependencies);
 

helpers/run-app.ts

@@ -34,24 +34,14 @@ export function runReflexApp(appPath: string, port: number) {
     "--frontend-port",
     port.toString(),
   ];
-  return createProcess("uv", commandArgs, {
+  return createProcess("poetry", commandArgs, {
     stdio: "inherit",
     cwd: appPath,
   });
 }
 
-export function runFastAPIApp(
-  appPath: string,
-  port: number,
-  template: TemplateType,
-) {
-  let commandArgs: string[];
-  if (template === "streaming") {
-    commandArgs = ["run", "dev"];
-  } else {
-    commandArgs = ["run", "fastapi", "dev", "--port", `${port}`];
-  }
-  return createProcess("uv", commandArgs, {
+export function runFastAPIApp(appPath: string, port: number) {
+  return createProcess("poetry", ["run", "dev"], {
     stdio: "inherit",
     cwd: appPath,
     env: { ...process.env, APP_PORT: `${port}` },
@@ -83,7 +73,7 @@ export async function runApp(
         : framework === "fastapi"
           ? runFastAPIApp
           : runTSApp;
-    await appRunner(appPath, port || defaultPort, template);
+    await appRunner(appPath, port || defaultPort);
   } catch (error) {
     console.error("Failed to run app:", error);
     throw error;

helpers/tools.ts

@@ -41,7 +41,7 @@ export const supportedTools: Tool[] = [
     dependencies: [
       {
         name: "llama-index-tools-google",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
       },
     ],
     supportedFrameworks: ["fastapi"],
@@ -62,7 +62,7 @@ export const supportedTools: Tool[] = [
     dependencies: [
       {
         name: "duckduckgo-search",
-        version: ">=6.3.5,<7.0.0",
+        version: "^6.3.5",
       },
     ],
     supportedFrameworks: ["fastapi"], // TODO: Re-enable this tool once the duck-duck-scrape TypeScript library works again
@@ -82,7 +82,7 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "llama-index-tools-wikipedia",
-        version: ">=0.3.0,<0.4.0",
+        version: "^0.3.0",
       },
     ],
     supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -102,11 +102,11 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "xhtml2pdf",
-        version: ">=0.2.14,<0.3.0",
+        version: "^0.2.14",
       },
       {
         name: "markdown",
-        version: ">=3.7.0,<4.0.0",
+        version: "^3.7",
       },
     ],
     type: ToolType.LOCAL,
@@ -124,7 +124,7 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "e2b_code_interpreter",
-        version: ">=1.1.1,<1.2.0",
+        version: "1.0.3",
       },
     ],
     supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -155,7 +155,7 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "e2b_code_interpreter",
-        version: ">=1.1.1,<1.2.0",
+        version: "1.0.3",
       },
     ],
     supportedFrameworks: ["fastapi", "express", "nextjs"],
@@ -184,7 +184,7 @@ For better results, you can specify the region parameter to get results from a s
       },
       {
         name: "jsonschema",
-        version: ">=4.22.0,<5.0.0",
+        version: "^4.22.0",
       },
       {
         name: "llama-index-tools-requests",
@@ -247,11 +247,11 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "pandas",
-        version: ">=2.2.3,<3.0.0",
+        version: "^2.2.3",
       },
       {
         name: "tabulate",
-        version: ">=0.9.0,<1.0.0",
+        version: "^0.9.0",
       },
     ],
   },

helpers/types.ts

@@ -24,8 +24,7 @@ export type TemplateType =
   | "community"
   | "llamapack"
   | "multiagent"
-  | "reflex"
-  | "llamaindexserver";
+  | "reflex";
 export type TemplateFramework = "nextjs" | "express" | "fastapi";
 export type TemplateUI = "html" | "shadcn";
 export type TemplateVectorDB =
@@ -56,9 +55,7 @@ export type TemplateUseCase =
   | "deep_research"
   | "form_filling"
   | "extractor"
-  | "contract_review"
-  | "agentic_rag"
-  | "artifacts";
+  | "contract_review";
 // Config for both file and folder
 export type FileSourceConfig =
   | {

helpers/typescript.ts

@@ -8,104 +8,42 @@ import { templatesDir } from "./dir";
 import { PackageManager } from "./get-pkg-manager";
 import { InstallTemplateArgs, ModelProvider, TemplateVectorDB } from "./types";
 
-const installLlamaIndexServerTemplate = async ({
-  root,
-  useCase,
-  vectorDb,
-}: Pick<InstallTemplateArgs, "root" | "useCase" | "vectorDb">) => {
-  if (!useCase) {
-    console.log(
-      red(
-        `There is no use case selected. Please pick a use case to use via --use-case flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  if (!vectorDb) {
-    console.log(
-      red(
-        `There is no vector db selected. Please pick a vector db to use via --vector-db flag.`,
-      ),
-    );
-    process.exit(1);
-  }
-
-  await copy("workflow.ts", path.join(root, "src", "app"), {
-    parents: true,
-    cwd: path.join(
-      templatesDir,
-      "components",
-      "workflows",
-      "typescript",
-      useCase,
-    ),
-  });
-
-  // copy workflow UI components to output/components folder
-  await copy("*", path.join(root, "components"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "ui", "workflows", useCase),
-  });
-
-  if (vectorDb === "llamacloud") {
-    await copy("generate.ts", path.join(root, "src"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "typescript",
-      ),
-    });
-
-    await copy("index.ts", path.join(root, "src", "app"), {
-      parents: true,
-      cwd: path.join(
-        templatesDir,
-        "components",
-        "vectordbs",
-        "llamaindexserver",
-        "llamacloud",
-        "typescript",
-      ),
-      rename: () => "data.ts",
-    });
-  }
-  // Copy README.md
-  await copy("README-template.md", path.join(root), {
-    parents: true,
-    cwd: path.join(
-      templatesDir,
-      "components",
-      "workflows",
-      "typescript",
-      useCase,
-    ),
-    rename: assetRelocator,
-  });
-};
-
-const installLegacyTSTemplate = async ({
+/**
+ * Install a LlamaIndex internal template to a given `root` directory.
+ */
+export const installTSTemplate = async ({
+  appName,
   root,
+  packageManager,
+  isOnline,
   template,
-  backend,
   framework,
   ui,
   vectorDb,
+  postInstallAction,
+  backend,
   observability,
   tools,
   dataSources,
   useLlamaParse,
   useCase,
   modelConfig,
-  relativeEngineDestPath,
-}: InstallTemplateArgs & {
-  backend: boolean;
-  relativeEngineDestPath: string;
-}) => {
+}: InstallTemplateArgs & { backend: boolean }) => {
+  console.log(bold(`Using ${packageManager}.`));
+
+  /**
+   * Copy the template files to the target directory.
+   */
+  console.log("\nInitializing project with template:", template, "\n");
+  const templatePath = path.join(templatesDir, "types", "streaming", framework);
+  const copySource = ["**"];
+
+  await copy(copySource, root, {
+    parents: true,
+    cwd: templatePath,
+    rename: assetRelocator,
+  });
+
   /**
    * If next.js is used, update its configuration if necessary
    */
@@ -160,6 +98,10 @@ const installLegacyTSTemplate = async ({
   }
 
   const compPath = path.join(templatesDir, "components");
+  const relativeEngineDestPath =
+    framework === "nextjs"
+      ? path.join("app", "api", "chat")
+      : path.join("src", "controllers");
   const enginePath = path.join(root, relativeEngineDestPath, "engine");
 
   // copy llamaindex code for TS templates
@@ -294,75 +236,6 @@ const installLegacyTSTemplate = async ({
     await fs.rm(path.join(root, "app", "api"), { recursive: true });
     await fs.rm(path.join(root, "config"), { recursive: true, force: true });
   }
-};
-
-/**
- * Install a LlamaIndex internal template to a given `root` directory.
- */
-export const installTSTemplate = async ({
-  appName,
-  root,
-  packageManager,
-  isOnline,
-  template,
-  framework,
-  ui,
-  vectorDb,
-  postInstallAction,
-  backend,
-  observability,
-  tools,
-  dataSources,
-  useLlamaParse,
-  useCase,
-  modelConfig,
-}: InstallTemplateArgs & { backend: boolean }) => {
-  console.log(bold(`Using ${packageManager}.`));
-
-  /**
-   * Copy the template files to the target directory.
-   */
-  console.log("\nInitializing project with template:", template, "\n");
-  const templatePath = path.join(templatesDir, "types", template, framework);
-  const copySource = ["**"];
-
-  await copy(copySource, root, {
-    parents: true,
-    cwd: templatePath,
-    rename: assetRelocator,
-  });
-
-  const relativeEngineDestPath =
-    framework === "nextjs"
-      ? path.join("app", "api", "chat")
-      : path.join("src", "controllers");
-
-  if (template === "llamaindexserver") {
-    await installLlamaIndexServerTemplate({
-      root,
-      useCase,
-      vectorDb,
-    });
-  } else {
-    await installLegacyTSTemplate({
-      appName,
-      root,
-      packageManager,
-      isOnline,
-      template,
-      backend,
-      framework,
-      ui,
-      vectorDb,
-      observability,
-      tools,
-      dataSources,
-      useLlamaParse,
-      useCase,
-      modelConfig,
-      relativeEngineDestPath,
-    });
-  }
 
   const packageJson = await updatePackageJson({
     root,
@@ -375,7 +248,6 @@ export const installTSTemplate = async ({
     vectorDb,
     backend,
     modelConfig,
-    template,
   });
 
   if (
@@ -390,27 +262,27 @@ const providerDependencies: {
   [key in ModelProvider]?: Record<string, string>;
 } = {
   openai: {
-    "@llamaindex/openai": "^0.2.0",
+    "@llamaindex/openai": "^0.1.52",
   },
   gemini: {
-    "@llamaindex/google": "^0.2.0",
+    "@llamaindex/google": "^0.0.7",
   },
   ollama: {
-    "@llamaindex/ollama": "^0.1.0",
+    "@llamaindex/ollama": "^0.0.40",
   },
   mistral: {
-    "@llamaindex/mistral": "^0.2.0",
+    "@llamaindex/mistral": "^0.0.5",
   },
   "azure-openai": {
-    "@llamaindex/openai": "^0.2.0",
+    "@llamaindex/openai": "^0.1.52",
   },
   groq: {
-    "@llamaindex/groq": "^0.0.61",
-    "@llamaindex/huggingface": "^0.1.0", // groq uses huggingface as default embedding model
+    "@llamaindex/groq": "^0.0.51",
+    "@llamaindex/huggingface": "^0.0.36", // groq uses huggingface as default embedding model
   },
   anthropic: {
-    "@llamaindex/anthropic": "^0.3.0",
-    "@llamaindex/huggingface": "^0.1.0", // anthropic uses huggingface as default embedding model
+    "@llamaindex/anthropic": "^0.1.0",
+    "@llamaindex/huggingface": "^0.0.36", // anthropic uses huggingface as default embedding model
   },
 };
 
@@ -459,7 +331,6 @@ async function updatePackageJson({
   vectorDb,
   backend,
   modelConfig,
-  template,
 }: Pick<
   InstallTemplateArgs,
   | "root"
@@ -470,7 +341,6 @@ async function updatePackageJson({
   | "observability"
   | "vectorDb"
   | "modelConfig"
-  | "template"
 > & {
   relativeEngineDestPath: string;
   backend: boolean;
@@ -482,7 +352,7 @@ async function updatePackageJson({
   packageJson.name = appName;
   packageJson.version = "0.1.0";
 
-  if (relativeEngineDestPath && template !== "llamaindexserver") {
+  if (relativeEngineDestPath) {
     // TODO: move script to {root}/scripts for all frameworks
     // add generate script if using context engine
     packageJson.scripts = {

helpers/validate-pkg.ts

@@ -1,3 +1,4 @@
+// eslint-disable-next-line import/no-extraneous-dependencies
 import validateProjectName from "validate-npm-package-name";
 
 export function validateNpmName(name: string): {

index.ts

@@ -1,3 +1,4 @@
+/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import { Command } from "commander";
 import fs from "fs";

llama-index-server/README.md

@@ -0,0 +1,55 @@
+# LlamaIndex Server
+
+LlamaIndexServer is a FastAPI application that allows you to quickly launch your workflow as an API server.
+
+## Installation
+
+```bash
+pip install llama-index-server
+```
+
+## Usage
+
+```python
+# main.py
+from llama_index.core.agent.workflow import AgentWorkflow
+from llama_index.core.workflow import Workflow
+from llama_index.core.tools import FunctionTool
+from llama_index.server import LlamaIndexServer
+
+
+# Define a factory function that returns a Workflow or AgentWorkflow
+def create_workflow() -> Workflow:
+    def fetch_weather(city: str) -> str:
+        return f"The weather in {city} is sunny"
+
+    return AgentWorkflow.from_tools(
+        tools=[
+            FunctionTool.from_defaults(
+                fn=fetch_weather,
+            )
+        ]
+    )
+
+
+# Create an API server the workflow
+app = LlamaIndexServer(
+    workflow_factory=create_workflow  # Supports Workflow or AgentWorkflow
+)
+```
+
+## Running the server
+
+- In the same directory as `main.py`, run the following command to start the server:
+
+  ```bash
+  fastapi dev
+  ```
+
+- Making a request to the server
+
+  ```bash
+  curl -X POST "http://localhost:8000/api/chat" -H "Content-Type: application/json" -d '{"message": "What is the weather in Tokyo?"}'
+  ```
+
+- See the API documentation at `http://localhost:8000/docs`

llama-index-server/llama_index/server/__init__.py

@@ -0,0 +1,3 @@
+from .server import LlamaIndexServer
+
+__all__ = ["LlamaIndexServer"]

llama-index-server/llama_index/server/api/callbacks/__init__.py

@@ -1,5 +1,4 @@
 from llama_index.server.api.callbacks.base import EventCallback
-from llama_index.server.api.callbacks.llamacloud import LlamaCloudFileDownload
 from llama_index.server.api.callbacks.source_nodes import SourceNodesFromToolCall
 from llama_index.server.api.callbacks.suggest_next_questions import (
     SuggestNextQuestions,
@@ -9,5 +8,4 @@ __all__ = [
     "EventCallback",
     "SourceNodesFromToolCall",
     "SuggestNextQuestions",
-    "LlamaCloudFileDownload",
 ]

llama-index-server/llama_index/server/api/models.py

@@ -1,14 +1,13 @@
 import logging
 import os
 from enum import Enum
-from typing import Any, Dict, List, Literal, Optional, Union
+from typing import Any, Dict, List, Optional
 
 from pydantic import BaseModel, Field, field_validator
 
 from llama_index.core.schema import NodeWithScore
 from llama_index.core.types import ChatMessage, MessageRole
 from llama_index.core.workflow import Event
-from llama_index.server.settings import server_settings
 
 logger = logging.getLogger("uvicorn")
 
@@ -23,7 +22,6 @@ class ChatConfig(BaseModel):
 class ChatAPIMessage(BaseModel):
     role: MessageRole
     content: str
-    annotations: Optional[List[Any]] = None
 
     def to_llamaindex_message(self) -> ChatMessage:
         return ChatMessage(role=self.role, content=self.content)
@@ -31,7 +29,6 @@ class ChatAPIMessage(BaseModel):
 
 class ChatRequest(BaseModel):
     messages: List[ChatAPIMessage]
-    data: Optional[Any] = None
     config: Optional[ChatConfig] = ChatConfig()
 
     @field_validator("messages")
@@ -101,11 +98,13 @@ class SourceNodes(BaseModel):
 
     @classmethod
     def get_url_from_metadata(
-        cls,
-        metadata: Dict[str, Any],
-        data_dir: Optional[str] = None,
+        cls, metadata: Dict[str, Any], data_dir: Optional[str] = None
     ) -> Optional[str]:
-        url_prefix = server_settings.file_server_url_prefix
+        url_prefix = os.getenv("FILESERVER_URL_PREFIX")
+        if not url_prefix:
+            logger.warning(
+                "Warning: FILESERVER_URL_PREFIX not set in environment variables. Can't use file server"
+            )
         if data_dir is None:
             data_dir = "data"
         file_name = metadata.get("file_name")
@@ -136,71 +135,3 @@ class SourceNodes(BaseModel):
         cls, source_nodes: List[NodeWithScore]
     ) -> List["SourceNodes"]:
         return [cls.from_source_node(node) for node in source_nodes]
-
-
-class ComponentDefinition(BaseModel):
-    type: str
-    code: str
-    filename: str
-
-
-class UIEvent(Event):
-    type: str
-    data: BaseModel
-
-    def to_response(self) -> dict:
-        return {
-            "type": self.type,
-            "data": self.data.model_dump(),
-        }
-
-
-class ArtifactType(str, Enum):
-    CODE = "code"
-    DOCUMENT = "document"
-
-
-class CodeArtifactData(BaseModel):
-    file_name: str
-    code: str
-    language: str
-
-
-class DocumentArtifactData(BaseModel):
-    title: str
-    content: str
-    type: Literal["markdown", "html"]
-
-
-class Artifact(BaseModel):
-    created_at: Optional[int] = None
-    type: ArtifactType
-    data: Union[CodeArtifactData, DocumentArtifactData]
-
-    @classmethod
-    def from_message(cls, message: ChatAPIMessage) -> Optional["Artifact"]:
-        if not message.annotations or not isinstance(message.annotations, list):
-            return None
-
-        for annotation in message.annotations:
-            if isinstance(annotation, dict) and annotation.get("type") == "artifact":
-                try:
-                    artifact = cls.model_validate(annotation.get("data"))
-                    return artifact
-                except Exception as e:
-                    logger.warning(
-                        f"Failed to parse artifact from annotation: {annotation}. Error: {e}"
-                    )
-
-        return None
-
-
-class ArtifactEvent(Event):
-    type: str = "artifact"
-    data: Artifact
-
-    def to_response(self) -> dict:
-        return {
-            "type": self.type,
-            "data": self.data.model_dump(),
-        }

llama-index-server/llama_index/server/api/routers/chat.py

@@ -1,28 +1,20 @@
 import asyncio
-import inspect
 import logging
-import os
 from typing import AsyncGenerator, Callable, Union
 
-from fastapi import APIRouter, BackgroundTasks, HTTPException
+from fastapi import APIRouter, HTTPException
 from fastapi.responses import StreamingResponse
 
-from llama_index.core.agent.workflow.workflow_events import (
-    AgentInput,
-    AgentSetup,
-    AgentStream,
-)
+from llama_index.core.agent.workflow.workflow_events import AgentStream
 from llama_index.core.workflow import StopEvent, Workflow
 from llama_index.server.api.callbacks import (
-    EventCallback,
-    LlamaCloudFileDownload,
     SourceNodesFromToolCall,
     SuggestNextQuestions,
 )
+from llama_index.server.api.callbacks.base import EventCallback
 from llama_index.server.api.callbacks.stream_handler import StreamHandler
 from llama_index.server.api.models import ChatRequest
 from llama_index.server.api.utils.vercel_stream import VercelStreamResponse
-from llama_index.server.services.llamacloud import LlamaCloudFileService
 
 
 def chat_router(
@@ -32,21 +24,13 @@ def chat_router(
     router = APIRouter(prefix="/chat")
 
     @router.post("")
-    async def chat(
-        request: ChatRequest,
-        background_tasks: BackgroundTasks,
-    ) -> StreamingResponse:
+    async def chat(request: ChatRequest) -> StreamingResponse:
         try:
             user_message = request.messages[-1].to_llamaindex_message()
             chat_history = [
                 message.to_llamaindex_message() for message in request.messages[:-1]
             ]
-            # detect if the workflow factory has chat_request as a parameter
-            factory_sig = inspect.signature(workflow_factory)
-            if "chat_request" in factory_sig.parameters:
-                workflow = workflow_factory(chat_request=request)
-            else:
-                workflow = workflow_factory()
+            workflow = workflow_factory()
             workflow_handler = workflow.run(
                 user_msg=user_message.content,
                 chat_history=chat_history,
@@ -54,7 +38,6 @@ def chat_router(
 
             callbacks: list[EventCallback] = [
                 SourceNodesFromToolCall(),
-                LlamaCloudFileDownload(background_tasks),
             ]
             if request.config and request.config.next_question_suggestions:
                 callbacks.append(SuggestNextQuestions(request))
@@ -70,28 +53,6 @@ def chat_router(
             logger.error(e)
             raise HTTPException(status_code=500, detail=str(e))
 
-    if LlamaCloudFileService.is_configured():
-
-        @router.get("/config/llamacloud")
-        async def chat_llama_cloud_config() -> dict:
-            if not os.getenv("LLAMA_CLOUD_API_KEY"):
-                raise HTTPException(
-                    status_code=500, detail="LlamaCloud API KEY is not configured"
-                )
-            projects = LlamaCloudFileService.get_all_projects_with_pipelines()
-            pipeline = os.getenv("LLAMA_CLOUD_INDEX_NAME")
-            project = os.getenv("LLAMA_CLOUD_PROJECT_NAME")
-            pipeline_config = None
-            if pipeline and project:
-                pipeline_config = {
-                    "pipeline": pipeline,
-                    "project": project,
-                }
-            return {
-                "projects": projects,
-                "pipeline": pipeline_config,
-            }
-
     return router
 
 
@@ -104,9 +65,7 @@ async def _stream_content(
         event: Union[AgentStream, StopEvent],
     ) -> AsyncGenerator[str, None]:
         if isinstance(event, AgentStream):
-            # Normally, if the stream is a tool call, the delta is always empty
-            # so it's not a text stream.
-            if len(event.tool_calls) == 0:
+            if event.delta.strip():  # Only yield non-empty deltas
                 yield event.delta
         elif isinstance(event, StopEvent):
             if isinstance(event.result, str):
@@ -115,11 +74,20 @@ async def _stream_content(
                 async for chunk in event.result:
                     if isinstance(chunk, str):
                         yield chunk
-                    elif hasattr(chunk, "delta") and chunk.delta:
+                    elif (
+                        hasattr(chunk, "delta") and chunk.delta.strip()
+                    ):  # Only yield non-empty deltas
                         yield chunk.delta
 
+    stream_started = False
     try:
         async for event in handler.stream_events():
+            if not stream_started:
+                # Start the stream with an empty message
+                stream_started = True
+                yield VercelStreamResponse.convert_text("")
+
+            # Handle different types of events
             if isinstance(event, (AgentStream, StopEvent)):
                 async for chunk in _text_stream(event):
                     handler.accumulate_text(chunk)
@@ -130,14 +98,12 @@ async def _stream_content(
                 event_response = event.to_response()
                 yield VercelStreamResponse.convert_data(event_response)
             else:
-                # Ignore unnecessary agent workflow events
-                if not isinstance(event, (AgentInput, AgentSetup)):
-                    yield VercelStreamResponse.convert_data(event.model_dump())
+                yield VercelStreamResponse.convert_data(event.model_dump())
 
     except asyncio.CancelledError:
         logger.warning("Client cancelled the request!")
         await handler.cancel_run()
     except Exception as e:
-        logger.error(f"Error in stream response: {e}", exc_info=True)
+        logger.error(f"Error in stream response: {e}")
         yield VercelStreamResponse.convert_error(str(e))
         await handler.cancel_run()

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 = "0.0.2"
 
 
 def download_chat_ui(

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

@@ -0,0 +1,134 @@
+import logging
+import os
+from typing import Any, Callable, Optional
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.staticfiles import StaticFiles
+
+from llama_index.core.workflow import Workflow
+from llama_index.server.api.routers.chat import chat_router
+from llama_index.server.chat_ui import download_chat_ui
+
+
+class LlamaIndexServer(FastAPI):
+    workflow_factory: Callable[..., Workflow]
+    api_prefix: str = "/api"
+    include_ui: Optional[bool]
+    verbose: bool = False
+    ui_path: str = ".ui"
+
+    def __init__(
+        self,
+        workflow_factory: Callable[..., Workflow],
+        logger: Optional[logging.Logger] = None,
+        use_default_routers: Optional[bool] = False,
+        env: Optional[str] = None,
+        include_ui: Optional[bool] = None,
+        verbose: bool = False,
+        *args: Any,
+        **kwargs: Any,
+    ):
+        """
+        Initialize the LlamaIndexServer.
+
+        Args:
+            workflow_factory: A factory function that creates a workflow instance for each request.
+            logger: The logger to use.
+            use_default_routers: Whether to use the default routers (chat, mount `data` and `output` directories).
+            env: The environment to run the server in.
+            include_ui: Whether to show an chat UI in the root path.
+            verbose: Whether to show verbose logs.
+        """
+        super().__init__(*args, **kwargs)
+
+        self.workflow_factory = workflow_factory
+        self.logger = logger or logging.getLogger("uvicorn")
+        self.verbose = verbose
+        self.include_ui = include_ui  # Store the explicitly passed value first
+
+        if use_default_routers:
+            self.add_default_routers()
+
+        if str(env).lower() == "dev":
+            self.allow_cors("*")
+            if self.include_ui is None:
+                self.include_ui = True
+        if self.include_ui is None:
+            self.include_ui = False
+
+        if self.include_ui:
+            self.mount_ui()
+
+    # Default routers
+    def add_default_routers(self) -> None:
+        self.add_chat_router()
+        self.mount_data_dir()
+        self.mount_output_dir()
+
+    def add_chat_router(self) -> None:
+        """
+        Add the chat router.
+        """
+        self.include_router(
+            chat_router(
+                self.workflow_factory,
+                self.logger,
+            ),
+            prefix=self.api_prefix,
+        )
+
+    def mount_ui(self) -> None:
+        """
+        Mount the UI.
+        """
+        # Check if the static folder exists
+        if self.include_ui:
+            if not os.path.exists(self.ui_path):
+                self.logger.warning(
+                    f"UI files not found, downloading UI to {self.ui_path}"
+                )
+                download_chat_ui(logger=self.logger, target_path=self.ui_path)
+            self._mount_static_files(directory=self.ui_path, path="/", html=True)
+
+    def mount_data_dir(self, data_dir: str = "data") -> None:
+        """
+        Mount the data directory.
+        """
+        self._mount_static_files(
+            directory=data_dir, path=f"{self.api_prefix}/files/data", html=True
+        )
+
+    def mount_output_dir(self, output_dir: str = "output") -> None:
+        """
+        Mount the output directory.
+        """
+        self._mount_static_files(
+            directory=output_dir, path=f"{self.api_prefix}/files/output", html=True
+        )
+
+    def _mount_static_files(
+        self, directory: str, path: str, html: bool = False
+    ) -> None:
+        """
+        Mount static files from a directory if it exists.
+        """
+        if os.path.exists(directory):
+            self.logger.info(f"Mounting static files '{directory}' at '{path}'")
+            self.mount(
+                path,
+                StaticFiles(directory=directory, check_dir=False, html=html),
+                name=f"{directory}-static",
+            )
+
+    def allow_cors(self, origin: str = "*") -> None:
+        """
+        Allow CORS for a specific origin.
+        """
+        self.add_middleware(
+            CORSMiddleware,
+            allow_origins=[origin],
+            allow_credentials=True,
+            allow_methods=["*"],
+            allow_headers=["*"],
+        )

llama-index-server/llama_index/server/services/file.py

@@ -5,7 +5,6 @@ import uuid
 from pathlib import Path
 from typing import List, Optional, Union
 
-from llama_index.server.settings import server_settings
 from pydantic import BaseModel, Field
 
 logger = logging.getLogger(__name__)
@@ -86,10 +85,20 @@ class FileService:
 
         logger.info(f"Saved file to {file_path}")
 
+        file_url_prefix = os.getenv("FILESERVER_URL_PREFIX")
+        if file_url_prefix is None:
+            logger.warning(
+                "FILESERVER_URL_PREFIX is not set. Some features may not work correctly."
+            )
+            file_url_prefix = "http://localhost:8000/api/files"
         file_size = os.path.getsize(file_path)
-        file_url = (
-            f"{server_settings.file_server_url_prefix}/{save_dir}/{new_file_name}"
+
+        file_url = os.path.join(
+            file_url_prefix,
+            save_dir,
+            new_file_name,
         )
+
         return DocumentFile(
             id=file_id,
             name=new_file_name,
@@ -100,15 +109,6 @@ class FileService:
             refs=None,
         )
 
-    @classmethod
-    def get_file_url(cls, file_name: str, save_dir: Optional[str] = None) -> str:
-        """
-        Get the URL of a file.
-        """
-        if save_dir is None:
-            save_dir = os.path.join("output", "uploaded")
-        return f"{server_settings.file_server_url_prefix}/{save_dir}/{file_name}"
-
 
 def _sanitize_file_name(file_name: str) -> str:
     """

llama-index-server/llama_index/server/tools/document_generator.py

@@ -99,11 +99,6 @@ HTML_TEMPLATE = """
 
 
 class DocumentGenerator:
-    def __init__(self, file_server_url_prefix: str):
-        if not file_server_url_prefix:
-            raise ValueError("file_server_url_prefix is required")
-        self.file_server_url_prefix = file_server_url_prefix
-
     @classmethod
     def _generate_html_content(cls, original_content: str) -> str:
         """
@@ -160,8 +155,9 @@ class DocumentGenerator:
             content=html_content,
         )
 
+    @classmethod
     def generate_document(
-        self, original_content: str, document_type: str, file_name: str
+        cls, original_content: str, document_type: str, file_name: str
     ) -> str:
         """
         To generate document as PDF or HTML file.
@@ -179,26 +175,24 @@ class DocumentGenerator:
                 f"Invalid document type: {document_type}. Must be 'pdf' or 'html'."
             )
         # Always generate html content first
-        html_content = self._generate_html_content(original_content)
+        html_content = cls._generate_html_content(original_content)
 
         # Based on the type of document, generate the corresponding file
         if doc_type == DocumentType.PDF:
-            content = self._generate_pdf(html_content)
+            content = cls._generate_pdf(html_content)
             file_extension = "pdf"
         elif doc_type == DocumentType.HTML:
-            content = BytesIO(self._generate_html(html_content).encode("utf-8"))
+            content = BytesIO(cls._generate_html(html_content).encode("utf-8"))
             file_extension = "html"
         else:
             raise ValueError(f"Unexpected document type: {document_type}")
 
-        file_name = self._validate_file_name(file_name)
+        file_name = cls._validate_file_name(file_name)
         file_path = os.path.join(OUTPUT_DIR, f"{file_name}.{file_extension}")
 
-        self._write_to_file(content, file_path)
+        cls._write_to_file(content, file_path)
 
-        return (
-            f"{self.file_server_url_prefix}/{OUTPUT_DIR}/{file_name}.{file_extension}"
-        )
+        return f"{os.getenv('FILESERVER_URL_PREFIX')}/{OUTPUT_DIR}/{file_name}.{file_extension}"
 
     @staticmethod
     def _write_to_file(content: BytesIO, file_path: str) -> None:

llama-index-server/llama_index/server/tools/interpreter.py

@@ -4,9 +4,10 @@ import os
 import uuid
 from typing import Any, List, Optional
 
+from pydantic import BaseModel
+
 from llama_index.core.tools import FunctionTool
 from llama_index.server.services.file import DocumentFile, FileService
-from pydantic import BaseModel
 
 logger = logging.getLogger("uvicorn")
 
@@ -33,24 +34,34 @@ class E2BCodeInterpreter:
 
     def __init__(
         self,
-        api_key: str,
+        api_key: Optional[str] = None,
+        filesever_url_prefix: Optional[str] = None,
         output_dir: Optional[str] = None,
         uploaded_files_dir: Optional[str] = None,
     ):
         """
         Args:
-            api_key: The API key for the E2B Code Interpreter.
+            api_key: The API key for the E2B Code Interpreter. If not provided, it will be read from the environment variable `E2B_API_KEY`.
+            filesever_url_prefix: The prefix for the file server or loaded from env: `FILESERVER_URL_PREFIX`, default is `/api/files`.
             output_dir: The directory for the output files. Default is `output/tools`.
             uploaded_files_dir: The directory for the files to be uploaded to the sandbox. Default is `output/uploaded`.
         """
         self._validate_package()
+        if api_key is None:
+            api_key = os.getenv("E2B_API_KEY")
+        if filesever_url_prefix is None:
+            filesever_url_prefix = os.getenv("FILESERVER_URL_PREFIX", "/api/files")
         if not api_key:
             raise ValueError(
-                "api_key is required to run code interpreter. Get it here: https://e2b.dev/docs/getting-started/api-key"
+                "E2B_API_KEY key is required to run code interpreter. Get it here: https://e2b.dev/docs/getting-started/api-key"
             )
+        if output_dir is not None:
+            self.output_dir = output_dir
+        if uploaded_files_dir is not None:
+            self.uploaded_files_dir = uploaded_files_dir
+        self.filesever_url_prefix = filesever_url_prefix
+        self.interpreter = None
         self.api_key = api_key
-        self.output_dir = output_dir or "output/tools"
-        self.uploaded_files_dir = uploaded_files_dir or "output/uploaded"
 
     @classmethod
     def _validate_package(cls) -> None:

llama-index-server/llama_index/server/utils/agent_tool.py

@@ -1,11 +1,11 @@
 import logging
 import uuid
 from abc import ABC, abstractmethod
-from typing import Any, AsyncGenerator, Optional
+from typing import Any, AsyncGenerator, Callable, Optional
 
 from pydantic import BaseModel, ConfigDict
 
-from llama_index.core.base.llms.types import ChatMessage, ChatResponse
+from llama_index.core.base.llms.types import ChatMessage, ChatResponse, MessageRole
 from llama_index.core.llms.function_calling import FunctionCallingLLM
 from llama_index.core.tools import (
     BaseTool,
@@ -19,7 +19,6 @@ from llama_index.core.agent.workflow.workflow_events import ToolCall, ToolCallRe
 
 logger = logging.getLogger("uvicorn")
 
-
 class ToolCallOutput(BaseModel):
     tool_call_id: str
     tool_output: ToolOutput
@@ -119,7 +118,11 @@ async def call_tools(
                 )
             )
         return [
-            await call_tool(ctx, tools_by_name[tool_calls[0].tool_name], tool_calls[0])
+            await call_tool(
+                ctx,
+                tools_by_name[tool_calls[0].tool_name],
+                tool_calls[0]
+            )
         ]
     # Multiple tool calls, show progress
     tool_call_outputs: list[ToolCallOutput] = []
@@ -147,7 +150,7 @@ async def call_tools(
                         raw_output={
                             "error": f"Tool {tool_call.tool_name} does not exist",
                         },
-                    ),
+                    )
                 )
             )
             continue
@@ -219,7 +222,6 @@ async def call_tool(
         tool_output=output,
     )
 
-
 async def _tool_call_generator(
     llm: FunctionCallingLLM,
     tools: list[BaseTool],

llama-index-server/pyproject.toml

@@ -0,0 +1,61 @@
+[build-system]
+build-backend = "poetry.core.masonry.api"
+requires = ["poetry-core"]
+
+[tool.codespell]
+check-filenames = true
+check-hidden = true
+# Feel free to un-skip examples, and experimental, you will just need to
+# work through many typos (--write-changes and --interactive will help)
+skip = "*.csv,*.html,*.json,*.jsonl,*.pdf,*.txt,*.ipynb"
+
+[tool.mypy]
+disallow_untyped_defs = true
+# Remove venv skip when integrated with pre-commit
+exclude = ["_static", "build", "examples", "notebooks", "venv"]
+ignore_missing_imports = true
+namespace_packages = true
+explicit_package_bases = true
+python_version = "3.10"
+
+[tool.poetry]
+authors = ["Your Name <you@example.com>"]
+description = "llama-index fastapi server"
+exclude = ["**/BUILD"]
+license = "MIT"
+name = "llama-index-server"
+packages = [{include = "llama_index/"}]
+readme = "README.md"
+version = "0.1.0"
+
+[tool.poetry.dependencies]
+python = ">=3.9,<4.0"
+fastapi = {extras = ["standard"], version = "^0.115.11"}
+cachetools = "^5.5.2"
+requests = "^2.32.3"
+llama-index-core = "^0.12.0"
+llama-index-readers-file = "^0.4.6"
+
+[tool.poetry.group.dev.dependencies]
+black = {extras = ["jupyter"], version = "<=23.9.1,>=23.7.0"}
+codespell = {extras = ["toml"], version = ">=v2.2.6"}
+e2b-code-interpreter = "^1.1.1"
+ipython = "8.10.0"
+jupyter = "^1.0.0"
+markdown = "^3.7"
+mypy = "1.15.0"
+pre-commit = "3.2.0"
+pylint = "2.15.10"
+pytest = "^8.3.5"
+pytest-asyncio = "^0.25.3"
+pytest-mock = "3.11.1"
+ruff = "0.0.292"
+tree-sitter-languages = "^1.8.0"
+types-Deprecated = ">=0.1.0"
+types-PyYAML = "^6.0.12.12"
+types-protobuf = "^4.24.0.4"
+types-redis = "4.5.5.0"
+types-requests = "2.28.11.8"  # TODO: unpin when mypy>0.991
+types-setuptools = "67.1.0.0"
+xhtml2pdf = "^0.2.17"
+pytest-cov = "^6.0.0"

llama-index-server/tests/api/test_event_stream.py

@@ -1,12 +1,10 @@
 import asyncio
 import logging
-from typing import Any, AsyncGenerator
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
 
 from llama_index.core.agent.workflow.workflow_events import AgentStream
-from llama_index.core.types import MessageRole
 from llama_index.core.workflow import StopEvent
 from llama_index.core.workflow.handler import WorkflowHandler
 from llama_index.server.api.models import ChatAPIMessage, ChatRequest
@@ -15,19 +13,17 @@ from llama_index.server.api.utils.vercel_stream import VercelStreamResponse
 
 
 @pytest.fixture()
-def logger() -> logging.Logger:
+def logger():
     return logging.getLogger("test")
 
 
 @pytest.fixture()
-def chat_request() -> ChatRequest:
-    return ChatRequest(
-        messages=[ChatAPIMessage(role=MessageRole.USER, content="test message")]
-    )
+def chat_request():
+    return ChatRequest(messages=[ChatAPIMessage(role="user", content="test message")])
 
 
 @pytest.fixture()
-def mock_workflow_handler() -> AsyncMock:
+def mock_workflow_handler():
     handler = AsyncMock(spec=WorkflowHandler)
     handler.accumulate_text = MagicMock()
     return handler
@@ -36,11 +32,8 @@ def mock_workflow_handler() -> AsyncMock:
 class TestEventStream:
     @pytest.mark.asyncio()
     async def test_stream_content_with_agent_stream(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.return_value = (
             self._mock_agent_stream_events()
@@ -55,17 +48,15 @@ class TestEventStream:
         ]
 
         # Assert
-        assert len(result) == 2
-        assert result[0] == VercelStreamResponse.convert_text("Hello")
-        assert result[1] == VercelStreamResponse.convert_text(" World")
+        assert len(result) == 3  # Empty start + 2 text chunks
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Hello")
+        assert result[2] == VercelStreamResponse.convert_text(" World")
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_stop_event_string(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.return_value = (
             self._mock_stop_event_string()
@@ -80,16 +71,14 @@ class TestEventStream:
         ]
 
         # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_text("Final answer")
+        assert len(result) == 2  # Empty start + result string
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Final answer")
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_stop_event_delta_objects(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.return_value = (
             self._mock_stop_event_delta_objects()
@@ -104,17 +93,15 @@ class TestEventStream:
         ]
 
         # Assert
-        assert len(result) == 2
-        assert result[0] == VercelStreamResponse.convert_text("Delta 1")
-        assert result[1] == VercelStreamResponse.convert_text("Delta 2")
+        assert len(result) == 3  # Empty start + 2 delta chunks
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_text("Delta 1")
+        assert result[2] == VercelStreamResponse.convert_text("Delta 2")
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_event_with_to_response(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.return_value = (
             self._mock_event_with_to_response()
@@ -129,16 +116,14 @@ class TestEventStream:
         ]
 
         # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_data({"event_type": "test"})
+        assert len(result) == 2  # Empty start + event with to_response
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_data({"event_type": "test"})
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_event_with_model_dump(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.return_value = (
             self._mock_event_with_model_dump()
@@ -153,19 +138,17 @@ class TestEventStream:
         ]
 
         # Assert
-        assert len(result) == 1
-        assert result[0] == VercelStreamResponse.convert_data(None)  # type: ignore
+        assert len(result) == 2  # Empty start + event with model_dump
+        assert result[0] == VercelStreamResponse.convert_text("")
+        assert result[1] == VercelStreamResponse.convert_data(None)
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_cancelled_error(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         mock_workflow_handler.stream_events.side_effect = asyncio.CancelledError()
-        logger.warning = MagicMock()  # type: ignore
+        logger.warning = MagicMock()
 
         # Execute
         result = [
@@ -182,15 +165,12 @@ class TestEventStream:
 
     @pytest.mark.asyncio()
     async def test_stream_content_with_exception(
-        self,
-        mock_workflow_handler: AsyncMock,
-        chat_request: ChatRequest,
-        logger: logging.Logger,
-    ) -> None:
+        self, mock_workflow_handler, chat_request, logger
+    ):
         # Setup
         error_message = "Test error"
         mock_workflow_handler.stream_events.side_effect = Exception(error_message)
-        logger.error = MagicMock()  # type: ignore
+        logger.error = MagicMock()
 
         # Execute
         result = [
@@ -206,7 +186,7 @@ class TestEventStream:
         mock_workflow_handler.cancel_run.assert_called_once()
         logger.error.assert_called_once()
 
-    async def _mock_agent_stream_events(self) -> AsyncGenerator[AgentStream, Any]:
+    async def _mock_agent_stream_events(self):
         yield AgentStream(
             delta="Hello", response="", current_agent_name="", tool_calls=[], raw=""
         )
@@ -214,9 +194,7 @@ class TestEventStream:
             delta=" World", response="", current_agent_name="", tool_calls=[], raw=""
         )
 
-    async def _mock_agent_stream_with_empty_deltas(
-        self,
-    ) -> AsyncGenerator[AgentStream, Any]:
+    async def _mock_agent_stream_with_empty_deltas(self):
         yield AgentStream(
             delta="   ",  # Empty delta with spaces - should be filtered
             response="",
@@ -239,30 +217,31 @@ class TestEventStream:
             raw="",
         )
 
-    async def _mock_stop_event_string(self) -> AsyncGenerator[StopEvent, Any]:
+    async def _mock_stop_event_string(self):
         yield StopEvent(result="Final answer")
 
-    async def _mock_stop_event_delta_objects(self) -> AsyncGenerator[StopEvent, Any]:
-        async def generator() -> AsyncGenerator[Any, Any]:
+    async def _mock_stop_event_delta_objects(self):
+        async def generator():
             # Create proper objects with delta attribute that can be serialized
             class ObjectWithDelta:
-                def __init__(self, delta_value: str) -> None:
+                def __init__(self, delta_value) -> None:
                     self.delta = delta_value
 
             yield ObjectWithDelta("Delta 1")
             yield ObjectWithDelta("Delta 2")
+            yield ObjectWithDelta("  ")  # Should be filtered out by strip check
 
         yield StopEvent(result=generator())
 
-    async def _mock_dict_event(self) -> AsyncGenerator[dict[Any, Any], Any]:
+    async def _mock_dict_event(self):
         yield {"key": "value"}
 
-    async def _mock_event_with_to_response(self) -> AsyncGenerator[Any, Any]:
+    async def _mock_event_with_to_response(self):
         event = MagicMock()
         event.to_response.return_value = {"event_type": "test"}
         yield event
 
-    async def _mock_event_with_model_dump(self) -> AsyncGenerator[Any, Any]:
+    async def _mock_event_with_model_dump(self):
         event = MagicMock()
         event.model_dump.return_value = {"name": "test_event"}
         # Override to_response to return None - this means convert_data(None) will be called

llama-index-server/tests/services/test_file_service.py

@@ -3,6 +3,7 @@ import uuid
 from unittest.mock import mock_open, patch
 
 import pytest
+
 from llama_index.server.services.file import FileService, _sanitize_file_name
 
 
@@ -53,7 +54,7 @@ class TestFileService:
         assert result.type == "txt"
         assert result.size == 11
         assert result.path == expected_path
-        assert result.url.endswith(expected_path.replace(os.path.sep, "/"))
+        assert result.url.endswith(expected_path)
         assert result.refs is None
 
     @patch("uuid.uuid4")
@@ -143,7 +144,7 @@ class TestFileService:
         test_uuid = "12345678-1234-5678-1234-567812345678"
         mock_uuid.return_value = uuid.UUID(test_uuid)
         mock_getsize.return_value = 11
-        mock_getenv.return_value = "/api/files"
+        mock_getenv.return_value = "https://custom-url.com/files"
 
         # Execute
         result = FileService.save_file(
@@ -157,8 +158,9 @@ class TestFileService:
         )
         mock_file_open.assert_called_once_with(expected_path, "wb")
         assert result.path == expected_path
-        # URL paths must use forward slashes, even on Windows
-        expected_url = f"/api/files/test_dir/test_{test_uuid}.txt"
+        expected_url = os.path.join(
+            "https://custom-url.com/files", "test_dir", f"test_{test_uuid}.txt"
+        )
         assert result.url == expected_url
 
     def test_save_file_no_extension(self):

llama-index-server/tests/test_llamaindex_server.py

@@ -0,0 +1,106 @@
+import pytest
+from httpx import ASGITransport, AsyncClient
+
+from llama_index.core.agent.workflow import AgentWorkflow
+from llama_index.core.llms import MockLLM
+from llama_index.server import LlamaIndexServer
+
+
+def fetch_weather(city: str) -> str:
+    """Fetch the weather for a given city."""
+    return f"The weather in {city} is sunny."
+
+
+def _agent_workflow() -> AgentWorkflow:
+    # Use MockLLM instead of default OpenAI
+    mock_llm = MockLLM()
+    return AgentWorkflow.from_tools_or_functions(
+        tools_or_functions=[fetch_weather],
+        verbose=True,
+        llm=mock_llm,
+    )
+
+
+@pytest.fixture()
+def server() -> LlamaIndexServer:
+    """Fixture to create a LlamaIndexServer instance."""
+    return LlamaIndexServer(
+        workflow_factory=_agent_workflow,
+        verbose=True,
+        use_default_routers=True,
+        mount_ui=False,
+        env="dev",
+    )
+
+
+@pytest.mark.asyncio()
+async def test_server_has_chat_route(server: LlamaIndexServer) -> None:
+    """Test that the server has the chat API route."""
+    chat_route_exists = any(route.path == "/api/chat" for route in server.routes)
+    assert chat_route_exists, "Chat API route not found in server routes"
+
+
+@pytest.mark.asyncio()
+async def test_server_swagger_docs(server: LlamaIndexServer) -> None:
+    """Test that the server serves Swagger UI docs."""
+    async with AsyncClient(
+        transport=ASGITransport(app=server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/docs")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]
+        assert "Swagger UI" in response.text
+
+
+@pytest.mark.asyncio()
+async def test_ui_is_downloaded(server: LlamaIndexServer) -> None:
+    """
+    Test if the UI is downloaded and mounted correctly.
+    """
+    import os
+    import shutil
+
+    # Clean up any existing static directory first
+    if os.path.exists(".ui"):
+        shutil.rmtree(".ui")
+
+    # Create a new server with UI enabled
+    ui_server = LlamaIndexServer(
+        workflow_factory=_agent_workflow,
+        verbose=True,
+        use_default_routers=True,
+        env="dev",
+        include_ui=True,
+    )
+
+    # Verify that static directory was created with index.html
+    assert os.path.exists("./.ui"), "Static directory was not created"
+    assert os.path.isdir("./.ui"), "Static path is not a directory"
+    assert os.path.exists("./.ui/index.html"), "index.html was not downloaded"
+
+    # Check if the UI is mounted and accessible
+    async with AsyncClient(
+        transport=ASGITransport(app=ui_server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]
+
+    # Clean up after test
+    shutil.rmtree("./.ui")
+
+
+@pytest.mark.asyncio()
+async def test_ui_is_accessible(server: LlamaIndexServer) -> None:
+    """
+    Test if the UI is accessible.
+    """
+    # Manually trigger UI mounting
+    server.mount_ui()
+
+    async with AsyncClient(
+        transport=ASGITransport(app=server), base_url="http://test"
+    ) as ac:
+        response = await ac.get("/")
+        assert response.status_code == 200
+        assert "text/html" in response.headers["content-type"]

llama-index-server/tests/tools/test_document_generator.py

@@ -1,7 +1,9 @@
+import os
 from io import BytesIO
 from unittest.mock import MagicMock, patch
 
 import pytest
+
 from llama_index.server.tools.document_generator import (
     OUTPUT_DIR,
     DocumentGenerator,
@@ -9,22 +11,25 @@ from llama_index.server.tools.document_generator import (
 
 
 class TestDocumentGenerator:
+    @pytest.fixture()
+    def env_setup(self):  # type: ignore
+        os.environ["FILESERVER_URL_PREFIX"] = "http://test-server"
+        yield
+        os.environ.pop("FILESERVER_URL_PREFIX", None)
+
     def test_validate_file_name(self) -> None:
         # Valid names
-        assert (
-            DocumentGenerator("/api/files")._validate_file_name("valid-name")
-            == "valid-name"
-        )
+        assert DocumentGenerator._validate_file_name("valid-name") == "valid-name"
 
         # Invalid names
         with pytest.raises(ValueError):
-            DocumentGenerator("/api/files")._validate_file_name("/invalid/path")
+            DocumentGenerator._validate_file_name("/invalid/path")
 
     @patch("os.makedirs")
     @patch("builtins.open")
     def test_write_to_file(self, mock_open, mock_makedirs):  # type: ignore
         content = BytesIO(b"test")
-        DocumentGenerator("/api/files")._write_to_file(content, "path/file.txt")
+        DocumentGenerator._write_to_file(content, "path/file.txt")
 
         mock_makedirs.assert_called_once()
         mock_open.assert_called_once()
@@ -37,13 +42,10 @@ class TestDocumentGenerator:
         mock_markdown.return_value = "<h1>Test</h1>"
 
         # Test HTML content generation
-        assert (
-            DocumentGenerator("/api/files")._generate_html_content("# Test")
-            == "<h1>Test</h1>"
-        )
+        assert DocumentGenerator._generate_html_content("# Test") == "<h1>Test</h1>"
 
         # Test full HTML generation
-        html = DocumentGenerator("/api/files")._generate_html("<h1>Test</h1>")
+        html = DocumentGenerator._generate_html("<h1>Test</h1>")
         assert "<!DOCTYPE html>" in html
         assert "<h1>Test</h1>" in html
 
@@ -51,14 +53,12 @@ class TestDocumentGenerator:
     def test_pdf_generation(self, mock_pisa):  # type: ignore
         # Success case
         mock_pisa.return_value = MagicMock(err=None)
-        assert isinstance(
-            DocumentGenerator("/api/files")._generate_pdf("test"), BytesIO
-        )
+        assert isinstance(DocumentGenerator._generate_pdf("test"), BytesIO)
 
         # Error case
         mock_pisa.return_value = MagicMock(err="Error")
         with pytest.raises(ValueError):
-            DocumentGenerator("/api/files")._generate_pdf("test")
+            DocumentGenerator._generate_pdf("test")
 
     @patch.multiple(
         DocumentGenerator,
@@ -69,21 +69,21 @@ class TestDocumentGenerator:
         _generate_pdf=MagicMock(return_value=BytesIO(b"pdf")),
         _write_to_file=MagicMock(),
     )
-    def test_generate_document(self):  # type: ignore
+    def test_generate_document(self, env_setup):  # type: ignore
         # HTML generation
-        url = DocumentGenerator("/api/files").generate_document(
-            "# Test", "html", "test-doc"
-        )
-        assert url == f"/api/files/{OUTPUT_DIR}/test-doc.html"
+        url = DocumentGenerator.generate_document("# Test", "html", "test-doc")
+        assert url == f"http://test-server/{OUTPUT_DIR}/test-doc.html"
 
         # PDF generation
-        url = DocumentGenerator("/api/files").generate_document(
-            "# Test", "pdf", "test-doc"
-        )
-        assert url == f"/api/files/{OUTPUT_DIR}/test-doc.pdf"
+        url = DocumentGenerator.generate_document("# Test", "pdf", "test-doc")
+        assert url == f"http://test-server/{OUTPUT_DIR}/test-doc.pdf"
 
         # Invalid type
         with pytest.raises(ValueError):
-            DocumentGenerator("/api/files").generate_document(
-                "# Test", "invalid", "test-doc"
-            )
+            DocumentGenerator.generate_document("# Test", "invalid", "test-doc")
+
+    def test_to_tool(self):  # type: ignore
+        tool = DocumentGenerator().to_tool()
+        # Check the function is correct
+        assert tool.fn == DocumentGenerator.generate_document
+        assert callable(tool.fn)

llama-index-server/tests/tools/test_interpreter.py

@@ -1,7 +1,9 @@
-from unittest.mock import MagicMock
+import os
+from unittest.mock import MagicMock, patch
 
 import pytest
 from e2b_code_interpreter.models import Execution, Logs
+
 from llama_index.server.tools.interpreter import E2BCodeInterpreter
 
 
@@ -18,9 +20,10 @@ class TestE2BCodeInterpreter:
     @pytest.fixture()
     def code_interpreter(self, sandbox):  # type: ignore
         """Create E2BCodeInterpreter that uses the mock Sandbox."""
-        interpreter = E2BCodeInterpreter(api_key="dummy_key")
-        interpreter.interpreter = sandbox
-        return interpreter
+        with patch.dict(os.environ, {"E2B_API_KEY": "dummy_key"}):
+            interpreter = E2BCodeInterpreter()
+            interpreter.interpreter = sandbox
+            return interpreter
 
     def test_interpret_success(self, code_interpreter, sandbox) -> None:  # type: ignore
         """Test successful code execution."""

package.json

@@ -1,52 +1,83 @@
 {
-  "name": "create-llama-monorepo",
-  "version": "1.0.0",
-  "private": true,
-  "description": "Monorepo for create-llama",
+  "name": "create-llama",
+  "version": "0.4.0",
+  "description": "Create LlamaIndex-powered apps with one command",
   "keywords": [
     "rag",
-    "llamaindex"
+    "llamaindex",
+    "next.js"
   ],
   "repository": {
     "type": "git",
-    "url": "https://github.com/run-llama/create-llama"
+    "url": "https://github.com/run-llama/create-llama",
+    "directory": "packages/create-llama"
   },
   "license": "MIT",
-  "workspaces": [
-    "packages/*"
+  "bin": {
+    "create-llama": "./dist/index.js"
+  },
+  "files": [
+    "dist"
   ],
   "scripts": {
-    "dev": "pnpm -r dev",
-    "build": "pnpm -r build",
-    "e2e": "pnpm -r e2e",
-    "lint": "eslint .",
+    "build": "bash ./scripts/build.sh",
+    "build:ncc": "pnpm run clean && ncc build ./index.ts -o ./dist/ --minify --no-cache --no-source-map-register",
+    "clean": "rimraf --glob ./dist ./templates/**/__pycache__ ./templates/**/node_modules ./templates/**/poetry.lock",
+    "dev": "ncc build ./index.ts -w -o dist/",
+    "e2e": "playwright test",
+    "e2e:python": "playwright test e2e/shared e2e/python",
+    "e2e:typescript": "playwright test e2e/shared e2e/typescript",
     "format": "prettier --ignore-unknown --cache --check .",
     "format:write": "prettier --ignore-unknown --write .",
+    "lint": "eslint . --ignore-pattern dist --ignore-pattern e2e/cache",
+    "new-snapshot": "pnpm run build && changeset version --snapshot",
+    "new-version": "pnpm run build && changeset version",
+    "pack-install": "bash ./scripts/pack.sh",
     "prepare": "husky",
-    "new-snapshot": "pnpm -r build && changeset version --snapshot",
-    "new-version": "pnpm -r build && changeset version",
-    "release": "pnpm -r build && changeset publish",
-    "release-snapshot": "pnpm -r build && changeset publish --tag snapshot"
+    "release": "pnpm run build && changeset publish",
+    "release-snapshot": "pnpm run build && changeset publish --tag snapshot"
+  },
+  "dependencies": {
+    "@types/async-retry": "1.4.2",
+    "@types/ci-info": "2.0.0",
+    "@types/cross-spawn": "6.0.0",
+    "@types/fs-extra": "11.0.4",
+    "@types/node": "^20.11.7",
+    "@types/prompts": "2.4.2",
+    "@types/tar": "6.1.5",
+    "@types/validate-npm-package-name": "3.0.0",
+    "async-retry": "1.3.1",
+    "async-sema": "3.0.1",
+    "ci-info": "github:watson/ci-info#f43f6a1cefff47fb361c88cf4b943fdbcaafe540",
+    "commander": "12.1.0",
+    "cross-spawn": "7.0.3",
+    "fast-glob": "3.3.1",
+    "fs-extra": "11.2.0",
+    "global-agent": "^3.0.0",
+    "got": "10.7.0",
+    "ollama": "^0.5.0",
+    "ora": "^8.0.1",
+    "picocolors": "1.0.0",
+    "prompts": "2.4.2",
+    "smol-toml": "^1.1.4",
+    "tar": "6.1.15",
+    "terminal-link": "^3.0.0",
+    "update-check": "1.5.4",
+    "validate-npm-package-name": "3.0.0",
+    "yaml": "2.4.1"
   },
   "devDependencies": {
     "@changesets/cli": "^2.27.1",
-    "bunchee": "6.4.0",
+    "@playwright/test": "^1.41.1",
+    "@vercel/ncc": "0.38.1",
+    "eslint": "^8.56.0",
+    "eslint-config-prettier": "^8.10.0",
     "husky": "^9.0.10",
-    "lint-staged": "^15.2.11",
-    "typescript-eslint": "^8.18.0",
-    "globals": "^15.12.0",
-    "eslint": "9.22.0",
-    "@eslint/js": "^9.25.0",
-    "eslint-config-next": "^15.1.0",
-    "eslint-config-prettier": "^9.1.0",
-    "eslint-plugin-react": "7.37.2",
-    "prettier": "^3.4.2",
-    "prettier-plugin-organize-imports": "^4.1.0",
-    "prettier-plugin-tailwindcss": "^0.6.11",
-    "typescript": "^5.7.3",
-    "@types/node": "^22.9.0",
-    "@types/react": "^19",
-    "@types/react-dom": "^19"
+    "prettier": "^3.2.5",
+    "prettier-plugin-organize-imports": "^3.2.4",
+    "rimraf": "^5.0.5",
+    "typescript": "^5.3.3",
+    "wait-port": "^1.1.0"
   },
   "packageManager": "pnpm@9.0.5",
   "engines": {

packages/create-llama/.gitignore

@@ -1,65 +0,0 @@
-# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
-
-# dependencies
-node_modules
-.pnp
-.pnpm-store
-.pnp.js
-
-# testing
-coverage
-.coverage
-
-# next.js
-.next/
-out/
-build
-
-# misc
-.DS_Store
-*.pem
-
-# debug
-npm-debug.log*
-yarn-debug.log*
-yarn-error.log*
-
-# local env files
-.env
-.env.local
-.env.development.local
-.env.test.local
-.env.production.local
-
-# build
-dist/
-lib/
-
-# e2e
-.cache
-test-results/
-playwright-report/
-blob-report/
-playwright/.cache/
-.tsbuildinfo
-e2e/cache
-
-# intellij
-**/.idea
-
-# Python
-.mypy_cache/
-venv/
-.venv/
-dist/
-.__pycache__
-__pycache__
-.python-version
-.ui
-
-# build artifacts
-create-llama-*.tgz
-
-# copied from root
-README.md
-LICENSE.md

packages/create-llama/helpers/uv.ts

@@ -1,42 +0,0 @@
-// Migrate poetry to uv
-import { execSync } from "child_process";
-import fs from "fs";
-import { red } from "picocolors";
-
-export function isUvAvailable(): boolean {
-  try {
-    execSync("uv --version", { stdio: "ignore" });
-    return true;
-  } catch (_) {}
-  return false;
-}
-
-export function tryUvSync(): boolean {
-  try {
-    console.log("Syncing environment with pyproject.toml...");
-    execSync(`uv sync`, {
-      stdio: "inherit",
-    });
-    return true;
-  } catch (_) {}
-  return false;
-}
-
-export function tryUvRun(command: string): boolean {
-  try {
-    // Use uv run <command>
-    execSync(`uv run ${command}`, { stdio: "inherit" });
-    return true;
-  } catch (error) {
-    console.error(red(`Failed to run ${command}. Error: ${error}`));
-    return false;
-  }
-}
-
-export function isHavingUvLockFile(): boolean {
-  try {
-    // Check if uv.lock exists in the current directory
-    return fs.existsSync("uv.lock");
-  } catch (_) {}
-  return false;
-}

packages/create-llama/package.json

@@ -1,76 +0,0 @@
-{
-  "name": "create-llama",
-  "version": "0.5.12",
-  "description": "Create LlamaIndex-powered apps with one command",
-  "keywords": [
-    "rag",
-    "llamaindex",
-    "next.js"
-  ],
-  "repository": {
-    "type": "git",
-    "url": "https://github.com/run-llama/create-llama",
-    "directory": "packages/create-llama"
-  },
-  "license": "MIT",
-  "bin": {
-    "create-llama": "./dist/index.js"
-  },
-  "files": [
-    "dist",
-    "README.md",
-    "LICENSE.md"
-  ],
-  "scripts": {
-    "copy": "cp -r ../../README.md ../../LICENSE.md .",
-    "build": "bash ./scripts/build.sh",
-    "build:ncc": "pnpm run clean && ncc build ./index.ts -o ./dist/ --minify --no-cache --no-source-map-register",
-    "postbuild": "pnpm run copy",
-    "clean": "rimraf --glob ./dist ./templates/**/__pycache__ ./templates/**/node_modules ./templates/**/poetry.lock",
-    "dev": "ncc build ./index.ts -w -o dist/",
-    "e2e": "playwright test",
-    "e2e:python": "playwright test e2e/shared e2e/python",
-    "e2e:typescript": "playwright test e2e/shared e2e/typescript",
-    "pack-install": "bash ./scripts/pack.sh"
-  },
-  "dependencies": {
-    "@types/async-retry": "1.4.2",
-    "@types/ci-info": "2.0.0",
-    "@types/cross-spawn": "6.0.0",
-    "@types/fs-extra": "11.0.4",
-    "@types/node": "^20.11.7",
-    "@types/prompts": "2.4.2",
-    "@types/tar": "6.1.5",
-    "@types/validate-npm-package-name": "3.0.0",
-    "async-retry": "1.3.1",
-    "async-sema": "3.0.1",
-    "ci-info": "github:watson/ci-info#f43f6a1cefff47fb361c88cf4b943fdbcaafe540",
-    "commander": "12.1.0",
-    "cross-spawn": "7.0.3",
-    "fast-glob": "3.3.1",
-    "fs-extra": "11.2.0",
-    "global-agent": "^3.0.0",
-    "got": "10.7.0",
-    "ollama": "^0.5.0",
-    "ora": "^8.0.1",
-    "picocolors": "1.0.0",
-    "prompts": "2.4.2",
-    "smol-toml": "^1.1.4",
-    "tar": "6.1.15",
-    "terminal-link": "^3.0.0",
-    "update-check": "1.5.4",
-    "validate-npm-package-name": "3.0.0",
-    "yaml": "2.4.1"
-  },
-  "devDependencies": {
-    "@playwright/test": "^1.41.1",
-    "@vercel/ncc": "0.38.1",
-    "rimraf": "^5.0.5",
-    "typescript": "^5.3.3",
-    "wait-port": "^1.1.0"
-  },
-  "packageManager": "pnpm@9.0.5",
-  "engines": {
-    "node": ">=16.14.0"
-  }
-}

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

@@ -1,33 +0,0 @@
-[project]
-name = "app"
-version = "0.1.0"
-description = ""
-authors = [ { name = "Marcus Schiesser", email = "mail@marcusschiesser.de" } ]
-readme = "README.md"
-requires-python = ">=3.11,<4.0"
-dependencies = [
-    "fastapi>=0.109.1",
-    "python-dotenv>=1.0.0",
-    "pydantic<2.10",
-    "llama-index>=0.12.1",
-    "cachetools>=5.3.3",
-    "reflex>=0.6.2.post1",
-]
-
-[project.scripts]
-generate = "app.engine.generate:generate_datasource"
-
-[project.optional-dependencies]
-dev = [
-    "mypy>=1.8.0",
-    "pytest>=8.3.5",
-    "pytest-asyncio>=0.25.3",
-    "docx2txt>=0.8",
-    "llama-index-llms-openai>=0.3.2",
-    "llama-index-embeddings-openai>=0.3.1",
-    "llama-index-agent-openai>=0.4.0",
-]
-
-[build-system]
-requires = [ "hatchling>=1.24" ]
-build-backend = "hatchling.build"

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

@@ -1,30 +0,0 @@
-[project]
-name = "app"
-version = "0.1.0"
-description = ""
-authors = [ { name = "Marcus Schiesser", email = "mail@marcusschiesser.de" } ]
-readme = "README.md"
-requires-python = ">=3.11,<4.0"
-dependencies = [
-    "fastapi>=0.109.1",
-    "uvicorn>=0.23.2",
-    "python-dotenv>=1.0.0",
-    "pydantic<2.10",
-    "llama-index>=0.12.1",
-    "cachetools>=5.3.3",
-    "reflex>=0.6.2.post1",
-]
-
-[project.scripts]
-generate = "app.engine.generate:generate_datasource"
-
-[project.optional-dependencies]
-dev = [
-    "mypy>=1.8.0",
-    "pytest>=8.3.5",
-    "pytest-asyncio>=0.25.3",
-]
-
-[build-system]
-requires = [ "hatchling>=1.24" ]
-build-backend = "hatchling.build"

packages/create-llama/templates/components/ui/workflows/artifacts/ui_event.jsx

@@ -1,137 +0,0 @@
-import { Badge } from "@/components/ui/badge";
-import { Card, CardContent, CardHeader, CardTitle } from "@/components/ui/card";
-import { Progress } from "@/components/ui/progress";
-import { Skeleton } from "@/components/ui/skeleton";
-import { cn } from "@/lib/utils";
-import { Markdown } from "@llamaindex/chat-ui/widgets";
-import { ListChecks, Loader2, Wand2 } from "lucide-react";
-import { useEffect, useState } from "react";
-
-const STAGE_META = {
-  plan: {
-    icon: ListChecks,
-    badgeText: "Step 1/2: Planning",
-    gradient: "from-blue-100 via-blue-50 to-white",
-    progress: 33,
-    iconBg: "bg-blue-100 text-blue-600",
-    badge: "bg-blue-100 text-blue-700",
-  },
-  generate: {
-    icon: Wand2,
-    badgeText: "Step 2/2: Generating",
-    gradient: "from-violet-100 via-violet-50 to-white",
-    progress: 66,
-    iconBg: "bg-violet-100 text-violet-600",
-    badge: "bg-violet-100 text-violet-700",
-  },
-};
-
-function ArtifactWorkflowCard({ event }) {
-  const [visible, setVisible] = useState(event?.state !== "completed");
-  const [fade, setFade] = useState(false);
-
-  useEffect(() => {
-    if (event?.state === "completed") {
-      setVisible(false);
-    } else {
-      setVisible(true);
-      setFade(false);
-    }
-  }, [event?.state]);
-
-  if (!event || !visible) return null;
-
-  const { state, requirement } = event;
-  const meta = STAGE_META[state];
-
-  if (!meta) return null;
-
-  return (
-    <div className="flex min-h-[180px] w-full items-center justify-center py-2">
-      <Card
-        className={cn(
-          "w-full rounded-xl shadow-md transition-all duration-500",
-          "border-0",
-          fade && "pointer-events-none opacity-0",
-          `bg-gradient-to-br ${meta.gradient}`,
-        )}
-        style={{
-          boxShadow:
-            "0 2px 12px 0 rgba(80, 80, 120, 0.08), 0 1px 3px 0 rgba(80, 80, 120, 0.04)",
-        }}
-      >
-        <CardHeader className="flex flex-row items-center gap-2 px-3 pb-1 pt-2">
-          <div
-            className={cn(
-              "flex items-center justify-center rounded-full p-1",
-              meta.iconBg,
-            )}
-          >
-            <meta.icon className="h-5 w-5" />
-          </div>
-          <CardTitle className="flex items-center gap-2 text-base font-semibold">
-            <Badge className={cn("ml-1", meta.badge, "px-2 py-0.5 text-xs")}>
-              {meta.badgeText}
-            </Badge>
-          </CardTitle>
-        </CardHeader>
-        <CardContent className="px-3 py-1">
-          {state === "plan" && (
-            <div className="flex flex-col items-center gap-2 py-2">
-              <Loader2 className="mb-1 h-6 w-6 animate-spin text-blue-400" />
-              <div className="text-center text-sm font-medium text-blue-900">
-                Analyzing your request...
-              </div>
-              <Skeleton className="mt-1 h-3 w-1/2 rounded-full" />
-            </div>
-          )}
-          {state === "generate" && (
-            <div className="flex flex-col gap-2 py-2">
-              <div className="flex items-center gap-1">
-                <Loader2 className="h-4 w-4 animate-spin text-violet-400" />
-                <span className="text-sm font-medium text-violet-900">
-                  Working on the requirement:
-                </span>
-              </div>
-              <div className="max-h-24 overflow-auto rounded-lg border border-violet-200 bg-violet-50 px-2 py-1 text-xs">
-                {requirement ? (
-                  <Markdown content={requirement} />
-                ) : (
-                  <span className="italic text-violet-400">
-                    No requirements available yet.
-                  </span>
-                )}
-              </div>
-            </div>
-          )}
-        </CardContent>
-        <div className="px-3 pb-2 pt-1">
-          <Progress
-            value={meta.progress}
-            className={cn(
-              "h-1 rounded-full bg-gray-200",
-              state === "plan" && "bg-blue-200",
-              state === "generate" && "bg-violet-200",
-            )}
-            indicatorClassName={cn(
-              "transition-all duration-500",
-              state === "plan" && "bg-blue-500",
-              state === "generate" && "bg-violet-500",
-            )}
-          />
-        </div>
-      </Card>
-    </div>
-  );
-}
-
-export default function Component({ events }) {
-  const aggregateEvents = () => {
-    if (!events || events.length === 0) return null;
-    return events[events.length - 1];
-  };
-
-  const event = aggregateEvents();
-
-  return <ArtifactWorkflowCard event={event} />;
-}

packages/create-llama/templates/components/ui/workflows/deep_research/ui_event.jsx

@@ -1,376 +0,0 @@
-import {
-  Accordion,
-  AccordionContent,
-  AccordionItem,
-  AccordionTrigger,
-} from "@/components/ui/accordion";
-import { Badge } from "@/components/ui/badge";
-import {
-  Card,
-  CardContent,
-  CardDescription,
-  CardFooter,
-  CardHeader,
-  CardTitle,
-} from "@/components/ui/card";
-import { Progress } from "@/components/ui/progress";
-import { Separator } from "@/components/ui/separator";
-import { cn } from "@/lib/utils";
-import { Markdown } from "@llamaindex/chat-ui/widgets";
-import {
-  AlertCircle,
-  Brain,
-  CheckCircle,
-  Clock,
-  Database,
-  Loader2,
-  MessageSquare,
-} from "lucide-react";
-import { useEffect, useState } from "react";
-
-export default function Component({ events }) {
-  const aggregateEvents = (events) => {
-    if (!events || events.length === 0)
-      return { retrieve: null, analyze: null, answers: [] };
-
-    // Initialize the result structure
-    const result = {
-      retrieve: null,
-      analyze: null,
-      answers: [],
-    };
-
-    // Process each event
-    events.forEach((event) => {
-      const { event: eventType, state, id, question, answer } = event;
-
-      if (eventType === "retrieve") {
-        // Update retrieve status
-        result.retrieve = { state };
-      } else if (eventType === "analyze") {
-        // Update analyze status
-        result.analyze = { state };
-      } else if (eventType === "answer" && id) {
-        // Find existing answer with the same id or create a new one
-        const existingAnswerIndex = result.answers.findIndex(
-          (a) => a.id === id,
-        );
-
-        if (existingAnswerIndex >= 0) {
-          // Update existing answer
-          result.answers[existingAnswerIndex] = {
-            ...result.answers[existingAnswerIndex],
-            state,
-            question: question || result.answers[existingAnswerIndex].question,
-            answer: answer || result.answers[existingAnswerIndex].answer,
-          };
-        } else {
-          // Add new answer
-          result.answers.push({
-            id,
-            state,
-            question,
-            answer,
-          });
-        }
-      }
-    });
-
-    return result;
-  };
-
-  const [aggregatedEvents, setAggregatedEvents] = useState({
-    retrieve: null,
-    analyze: null,
-    answers: [],
-  });
-
-  useEffect(() => {
-    setAggregatedEvents(aggregateEvents(events));
-  }, [events]);
-
-  const { retrieve, analyze, answers } = aggregatedEvents;
-
-  // Helper function to get status icon
-  const getStatusIcon = (state) => {
-    switch (state) {
-      case "pending":
-        return <Clock className="h-4 w-4 text-gray-400" />;
-      case "inprogress":
-        return <Loader2 className="h-4 w-4 animate-spin text-blue-500" />;
-      case "done":
-        return <CheckCircle className="h-4 w-4 text-green-500" />;
-      case "error":
-        return <AlertCircle className="h-4 w-4 text-red-500" />;
-      default:
-        return null;
-    }
-  };
-
-  // Helper function to get status text
-  const getStatusText = (state) => {
-    switch (state) {
-      case "pending":
-        return "Pending";
-      case "inprogress":
-        return "In Progress";
-      case "done":
-        return "Complete";
-      case "error":
-        return "Error";
-      default:
-        return "";
-    }
-  };
-
-  // Helper function to get status color class
-  const getStatusColorClass = (state) => {
-    switch (state) {
-      case "pending":
-        return "bg-gray-200";
-      case "inprogress":
-        return "bg-blue-500";
-      case "done":
-        return "bg-green-500";
-      case "error":
-        return "bg-red-500";
-      default:
-        return "bg-gray-200";
-    }
-  };
-
-  return (
-    <div className="mx-auto w-full max-w-4xl space-y-6 p-4">
-      {/* Header */}
-      <div className="mb-6 flex items-center justify-between">
-        <h1 className="text-2xl font-bold">DeepResearch Workflow</h1>
-        <div className="flex items-center space-x-2">
-          <Badge
-            variant={retrieve?.state === "done" ? "default" : "outline"}
-            className={cn(
-              "transition-all",
-              retrieve?.state === "done" ? "bg-green-500" : "",
-            )}
-          >
-            Retrieve
-          </Badge>
-          <Separator className="h-4 w-px bg-gray-300" orientation="vertical" />
-          <Badge
-            variant={analyze?.state === "done" ? "default" : "outline"}
-            className={cn(
-              "transition-all",
-              analyze?.state === "done" ? "bg-green-500" : "",
-            )}
-          >
-            Analyze
-          </Badge>
-          <Separator className="h-4 w-px bg-gray-300" orientation="vertical" />
-          <Badge
-            variant={
-              answers.length > 0 && answers.every((a) => a.state === "done")
-                ? "default"
-                : "outline"
-            }
-            className={cn(
-              "transition-all",
-              answers.length > 0 && answers.every((a) => a.state === "done")
-                ? "bg-green-500"
-                : "",
-            )}
-          >
-            Answer
-          </Badge>
-        </div>
-      </div>
-
-      {/* Retrieve Panel */}
-      <Card
-        className={cn(
-          "border-2 transition-all duration-300",
-          retrieve?.state === "inprogress"
-            ? "border-blue-500 shadow-lg shadow-blue-100"
-            : retrieve?.state === "done"
-              ? "border-green-500"
-              : retrieve?.state === "error"
-                ? "border-red-500"
-                : "border-gray-200",
-        )}
-      >
-        <CardHeader>
-          <div className="flex items-center justify-between">
-            <div className="flex items-center space-x-2">
-              <Database className="h-5 w-5 text-gray-700" />
-              <CardTitle>Retrieve Information</CardTitle>
-            </div>
-            <Badge
-              variant="outline"
-              className={cn(
-                "flex items-center space-x-1",
-                retrieve?.state === "inprogress"
-                  ? "text-blue-500"
-                  : retrieve?.state === "done"
-                    ? "text-green-500"
-                    : retrieve?.state === "error"
-                      ? "text-red-500"
-                      : "text-gray-500",
-              )}
-            >
-              {getStatusIcon(retrieve?.state)}
-              <span>{getStatusText(retrieve?.state)}</span>
-            </Badge>
-          </div>
-          <CardDescription>
-            Retrieving relevant information from the knowledge base
-          </CardDescription>
-        </CardHeader>
-      </Card>
-
-      {/* Analyze Panel */}
-      {retrieve?.state === "done" && (
-        <Card
-          className={cn(
-            "border-2 transition-all duration-300",
-            analyze?.state === "inprogress"
-              ? "border-blue-500 shadow-lg shadow-blue-100"
-              : analyze?.state === "done"
-                ? "border-green-500"
-                : analyze?.state === "error"
-                  ? "border-red-500"
-                  : "border-gray-200",
-          )}
-        >
-          <CardHeader>
-            <div className="flex items-center justify-between">
-              <div className="flex items-center space-x-2">
-                <Brain className="h-5 w-5 text-gray-700" />
-                <CardTitle>Analyze Information</CardTitle>
-              </div>
-              <Badge
-                variant="outline"
-                className={cn(
-                  "flex items-center space-x-1",
-                  analyze?.state === "inprogress"
-                    ? "text-blue-500"
-                    : analyze?.state === "done"
-                      ? "text-green-500"
-                      : analyze?.state === "error"
-                        ? "text-red-500"
-                        : "text-gray-500",
-                )}
-              >
-                {getStatusIcon(analyze?.state)}
-                <span>{getStatusText(analyze?.state)}</span>
-              </Badge>
-            </div>
-            <CardDescription>
-              Analyzing retrieved information and generating questions
-            </CardDescription>
-          </CardHeader>
-        </Card>
-      )}
-
-      {/* Answer Panel */}
-      {analyze?.state === "done" && answers.length > 0 && (
-        <Card>
-          <CardHeader>
-            <div className="flex items-center space-x-2">
-              <MessageSquare className="h-5 w-5 text-gray-700" />
-              <CardTitle>Answers</CardTitle>
-            </div>
-            <CardDescription>
-              Detailed answers to the generated questions
-            </CardDescription>
-          </CardHeader>
-          <CardContent>
-            <Accordion type="multiple" className="w-full">
-              {answers.map((answer, index) => (
-                <AccordionItem
-                  key={answer.id}
-                  value={answer.id}
-                  className={cn(
-                    "mb-4 overflow-hidden rounded-lg border",
-                    answer.state === "inprogress"
-                      ? "border-blue-500 shadow-sm shadow-blue-100"
-                      : answer.state === "done"
-                        ? "border-green-100"
-                        : answer.state === "error"
-                          ? "border-red-100"
-                          : "border-gray-200",
-                  )}
-                >
-                  <AccordionTrigger className="px-4 py-3 hover:bg-gray-50">
-                    <div className="flex items-center space-x-3 text-left">
-                      <Badge className="shrink-0 bg-gray-700 text-white">
-                        Q{index + 1}
-                      </Badge>
-                      <div className="flex-1">
-                        <p className="font-medium">{answer.question}</p>
-                      </div>
-                      <Badge
-                        variant="outline"
-                        className={cn(
-                          "ml-auto flex shrink-0 items-center space-x-1",
-                          answer.state === "inprogress"
-                            ? "text-blue-500"
-                            : answer.state === "done"
-                              ? "text-green-500"
-                              : answer.state === "error"
-                                ? "text-red-500"
-                                : "text-gray-500",
-                        )}
-                      >
-                        {getStatusIcon(answer.state)}
-                        <span>{getStatusText(answer.state)}</span>
-                      </Badge>
-                    </div>
-                  </AccordionTrigger>
-                  <AccordionContent className="px-4 pb-4 pt-1">
-                    <div
-                      className={cn(
-                        "rounded-md p-3",
-                        answer.state === "done"
-                          ? "bg-green-50"
-                          : answer.state === "inprogress"
-                            ? "bg-blue-50"
-                            : "bg-gray-50",
-                      )}
-                    >
-                      {answer.answer ? (
-                        <Markdown content={answer.answer} />
-                      ) : (
-                        <div className="flex items-center justify-center p-4 text-gray-500">
-                          {answer.state === "inprogress" ? (
-                            <div className="flex items-center space-x-2">
-                              <Loader2 className="h-4 w-4 animate-spin" />
-                              <span>Generating answer...</span>
-                            </div>
-                          ) : (
-                            <span>Waiting for answer</span>
-                          )}
-                        </div>
-                      )}
-                    </div>
-                  </AccordionContent>
-                </AccordionItem>
-              ))}
-            </Accordion>
-          </CardContent>
-          <CardFooter className="flex justify-between">
-            <div className="text-sm text-gray-500">
-              {answers.filter((a) => a.state === "done").length} of{" "}
-              {answers.length} questions answered
-            </div>
-            <Progress
-              value={
-                (answers.filter((a) => a.state === "done").length /
-                  answers.length) *
-                100
-              }
-              className="h-2 w-1/3 bg-gray-200"
-            />
-          </CardFooter>
-        </Card>
-      )}
-    </div>
-  );
-}

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

@@ -1,31 +0,0 @@
-# flake8: noqa: E402
-
-from dotenv import load_dotenv
-
-load_dotenv()
-
-import logging
-
-from app.index import get_index
-from app.settings import init_settings
-from llama_index.server.services.llamacloud.generate import (
-    load_to_llamacloud,
-)
-
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger()
-
-
-def generate_datasource():
-    init_settings()
-    logger.info("Generate index for the provided data")
-
-    index = get_index(create_if_missing=True)
-    if index is None:
-        raise ValueError("Index not found and could not be created")
-
-    load_to_llamacloud(index, logger=logger)
-
-
-if __name__ == "__main__":
-    generate_datasource()

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

@@ -1,7 +0,0 @@
-from llama_index.server.services.llamacloud import (
-    LlamaCloudIndex,
-    get_client,
-    get_index,
-)
-
-__all__ = ["LlamaCloudIndex", "get_client", "get_index"]