Release 0.2.9 (#699)

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

.eslintrc

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

.github/workflows/e2e.yml

@@ -1,12 +1,15 @@
-name: E2E Tests
+name: E2E Tests for create-llama package
 on:
   push:
     branches: [main]
+    paths-ignore:
+      - "python/llama-index-server/**"
+      - ".github/workflows/*llama_index_server.yml"
   pull_request:
     branches: [main]
-
-env:
-  POETRY_VERSION: "1.6.1"
+    paths-ignore:
+      - "python/llama-index-server/**"
+      - ".github/workflows/*llama_index_server.yml"
 
 jobs:
   e2e-python:
@@ -19,7 +22,7 @@ jobs:
         python-version: ["3.11"]
         os: [macos-latest, windows-latest, ubuntu-22.04]
         frameworks: ["fastapi"]
-        datasources: ["--no-files", "--example-file", "--llamacloud"]
+        vectordbs: ["none", "llamacloud"]
     defaults:
       run:
         shell: bash
@@ -32,10 +35,10 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
 
-      - name: Install Poetry
-        uses: snok/install-poetry@v1
-        with:
-          version: ${{ env.POETRY_VERSION }}
+      - name: Install uv
+        run: curl -LsSf https://astral.sh/uv/install.sh | sh
+      - name: Add uv to PATH # Ensure uv is available in subsequent steps
+        run: echo "$HOME/.cargo/bin" >> $GITHUB_PATH
 
       - uses: pnpm/action-setup@v3
 
@@ -50,15 +53,24 @@ jobs:
 
       - name: Install Playwright Browsers
         run: pnpm exec playwright install --with-deps
-        working-directory: .
+        working-directory: packages/create-llama
 
       - name: Build create-llama
         run: pnpm run build
-        working-directory: .
+        working-directory: packages/create-llama
 
       - name: Install
         run: pnpm run pack-install
-        working-directory: .
+        working-directory: packages/create-llama
+
+      - name: Build and store server package
+        run: |
+          pnpm run build
+          wheel_file=$(ls dist/*.whl | head -n 1)
+          mkdir -p "${{ runner.temp }}"
+          cp "$wheel_file" "${{ runner.temp }}/"
+          echo "SERVER_PACKAGE_PATH=${{ runner.temp }}/$(basename "$wheel_file")" >> $GITHUB_ENV
+        working-directory: python/llama-index-server
 
       - name: Run Playwright tests for Python
         run: pnpm run e2e:python
@@ -66,16 +78,17 @@ jobs:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
           FRAMEWORK: ${{ matrix.frameworks }}
-          DATASOURCE: ${{ matrix.datasources }}
+          VECTORDB: ${{ matrix.vectordbs }}
           PYTHONIOENCODING: utf-8
           PYTHONLEGACYWINDOWSSTDIO: utf-8
-        working-directory: .
+          SERVER_PACKAGE_PATH: ${{ env.SERVER_PACKAGE_PATH }}
+        working-directory: packages/create-llama
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
-          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}
-          path: ./playwright-report/
+          name: playwright-report-python-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.vectordbs }}
+          path: packages/create-llama/playwright-report/
           overwrite: true
           retention-days: 30
 
@@ -85,11 +98,10 @@ jobs:
     strategy:
       fail-fast: true
       matrix:
-        node-version: [18, 20]
-        python-version: ["3.11"]
+        node-version: [22]
         os: [macos-latest, windows-latest, ubuntu-22.04]
         frameworks: ["nextjs"]
-        datasources: ["--no-files", "--example-file", "--llamacloud"]
+        vectordbs: ["none", "llamacloud"]
     defaults:
       run:
         shell: bash
@@ -97,16 +109,6 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - name: Set up python ${{ matrix.python-version }}
-        uses: actions/setup-python@v5
-        with:
-          python-version: ${{ matrix.python-version }}
-
-      - name: Install Poetry
-        uses: snok/install-poetry@v1
-        with:
-          version: ${{ env.POETRY_VERSION }}
-
       - uses: pnpm/action-setup@v3
 
       - name: Setup Node.js ${{ matrix.node-version }}
@@ -120,29 +122,46 @@ jobs:
 
       - name: Install Playwright Browsers
         run: pnpm exec playwright install --with-deps
-        working-directory: .
+        working-directory: packages/create-llama
 
       - name: Build create-llama
         run: pnpm run build
-        working-directory: .
+        working-directory: packages/create-llama
 
       - name: Install
         run: pnpm run pack-install
-        working-directory: .
+        working-directory: packages/create-llama
+
+      - name: Build server
+        run: pnpm run build
+        working-directory: packages/server
+
+      - name: Pack @llamaindex/server package
+        run: |
+          pnpm pack --pack-destination "${{ runner.temp }}"
+          if [ "${{ runner.os }}" == "Windows" ]; then
+            file=$(find "${{ runner.temp }}" -name "llamaindex-server-*.tgz" | head -n 1)
+            mv "$file" "${{ runner.temp }}/llamaindex-server.tgz"
+          else
+            mv ${{ runner.temp }}/llamaindex-server-*.tgz ${{ runner.temp }}/llamaindex-server.tgz
+          fi
+        working-directory: packages/server
 
       - name: Run Playwright tests for TypeScript
-        run: pnpm run e2e:typescript
+        run: |
+          pnpm run e2e:ts
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           LLAMA_CLOUD_API_KEY: ${{ secrets.LLAMA_CLOUD_API_KEY }}
           FRAMEWORK: ${{ matrix.frameworks }}
-          DATASOURCE: ${{ matrix.datasources }}
-        working-directory: .
+          VECTORDB: ${{ matrix.vectordbs }}
+          SERVER_PACKAGE_PATH: ${{ runner.temp }}/llamaindex-server.tgz
+        working-directory: packages/create-llama
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
-          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.datasources }}-node${{ matrix.node-version }}
-          path: ./playwright-report/
+          name: playwright-report-typescript-${{ matrix.os }}-${{ matrix.frameworks }}-${{ matrix.vectordbs}}-node${{ matrix.node-version }}
+          path: packages/create-llama/playwright-report/
           overwrite: true
           retention-days: 30

.github/workflows/lint_on_push_or_pull.yml

@@ -16,6 +16,16 @@ jobs:
 
       - uses: pnpm/action-setup@v3
 
+      - name: Setup Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -31,12 +41,21 @@ jobs:
       - name: Run Prettier
         run: pnpm run format
 
+      - name: Run build
+        run: pnpm run build
+
+      - name: Run Typecheck for examples
+        run: pnpm run typecheck
+        working-directory: packages/server/examples
+
       - name: Run Python format check
         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.yml

@@ -17,6 +17,11 @@ jobs:
 
       - uses: pnpm/action-setup@v3
 
+      - name: Setup Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
       - name: Install uv
         uses: astral-sh/setup-uv@v3
 
@@ -51,8 +56,12 @@ jobs:
         with:
           commit: Release ${{ steps.get-changeset-status.outputs.new-version }}
           title: Release ${{ steps.get-changeset-status.outputs.new-version }}
+          # bump versions
+          version: pnpm new-version
           # build package and call changeset publish
           publish: pnpm release
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           NPM_TOKEN: ${{ secrets.NPM_TOKEN }}
+          PYPI_TOKEN: ${{ secrets.PYPI_TOKEN }}
+          UV_PUBLISH_TOKEN: ${{ secrets.PYPI_TOKEN }}

.github/workflows/release_llama_index_server.yml

@@ -1,130 +0,0 @@
-name: Release llama-index-server
-
-on:
-  push:
-    branches:
-      - main
-    paths:
-      - "llama-index-server/**"
-      - ".github/workflows/release_llama_index_server.yml"
-  pull_request:
-    types:
-      - closed
-
-concurrency: ${{ github.workflow }}-${{ github.ref }}
-
-jobs:
-  release:
-    name: Create Release PR
-    runs-on: ubuntu-latest
-    defaults:
-      run:
-        working-directory: ./llama-index-server
-    if: |
-      github.event_name == 'push' && 
-      !startsWith(github.ref, 'refs/heads/release/llama-index-server-v')
-
-    steps:
-      - name: Checkout Repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      - 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
-        run: poetry install
-
-      - name: Setup Git
-        run: |
-          git config --global user.email "github-actions[bot]@users.noreply.github.com"
-          git config --global user.name "github-actions[bot]"
-
-      - name: Bump patch version
-        run: |
-          poetry version patch
-          git add pyproject.toml
-          git commit -m "chore(release): bump version to $(poetry version -s)"
-
-      - name: Get current version
-        id: get_version
-        run: |
-          version=$(poetry version -s)
-          echo "current_version=${version}" >> "$GITHUB_OUTPUT"
-
-      - name: Create Release PR
-        uses: peter-evans/create-pull-request@v6
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          commit-message: "Release: llama-index-server v${{ steps.get_version.outputs.current_version }}"
-          title: "Release: llama-index-server v${{ steps.get_version.outputs.current_version }}"
-          body: |
-            This PR was automatically created to release a new version of the llama-index-server package.
-
-            Version: ${{ steps.get_version.outputs.current_version }}
-
-            Please review the changes and merge to trigger the release.
-          branch: release/llama-index-server-v${{ steps.get_version.outputs.current_version }}
-          base: main
-          labels: release, llama-index-server
-
-  publish:
-    name: Publish to PyPI
-    runs-on: ubuntu-latest
-    defaults:
-      run:
-        working-directory: ./llama-index-server
-    if: |
-      github.event_name == 'pull_request' && 
-      github.event.pull_request.merged == true && 
-      startsWith(github.event.pull_request.title, 'Release: llama-index-server') &&
-      startsWith(github.event.pull_request.head.ref, 'release/llama-index-server-v')
-
-    steps:
-      - name: Checkout Repository
-        uses: actions/checkout@v4
-
-      - 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
-        run: poetry install
-
-      - name: Get current version
-        id: get_version
-        run: |
-          version=$(poetry version -s)
-          echo "current_version=${version}" >> "$GITHUB_OUTPUT"
-
-      - name: Build and publish to PyPI
-        uses: JRubics/poetry-publish@v2.1
-        with:
-          python_version: "3.11"
-          pypi_token: ${{ secrets.PYPI_TOKEN }}
-          package_directory: "llama-index-server"
-          poetry_install_options: "--without dev"
-
-      - name: Create GitHub Release
-        uses: softprops/action-gh-release@v2
-        with:
-          tag_name: llama-index-server-v${{ steps.get_version.outputs.current_version }}
-          name: "llama-index-server v${{ steps.get_version.outputs.current_version }}"
-          body: |
-            Release of llama-index-server v${{ steps.get_version.outputs.current_version }}
-          draft: false
-          prerelease: false
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/test_llama_index_server.yml

@@ -4,8 +4,8 @@ on:
   pull_request:
 
 env:
-  POETRY_VERSION: "1.8.3"
   PYTHON_VERSION: "3.9"
+  UI_TEST: "true"
 
 jobs:
   unit-test:
@@ -13,99 +13,124 @@ jobs:
     runs-on: ${{ matrix.os }}
     defaults:
       run:
-        working-directory: llama-index-server
+        working-directory: python/llama-index-server
     strategy:
       matrix:
         os: [ubuntu-latest, windows-latest]
         python-version: ["3.9"]
     steps:
       - uses: actions/checkout@v4
+      - uses: pnpm/action-setup@v3
 
-      - name: Install Poetry
-        run: pipx install poetry==${{ env.POETRY_VERSION }}
-
-      - name: Set up python ${{ matrix.python-version }}
+      - name: Setup Python
         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 uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
 
       - name: Install dependencies
         shell: bash
-        run: poetry install --with dev
+        run: pnpm install && pnpm build
 
       - name: Run unit tests
         shell: bash
-        run: |
-          poetry run pytest tests
+        run: uv run pytest tests
 
   type-check:
     name: Type Check
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: llama-index-server
+        working-directory: python/llama-index-server
     steps:
       - uses: actions/checkout@v4
+      - uses: pnpm/action-setup@v3
 
-      - name: Install Poetry
-        run: pipx install poetry==${{ env.POETRY_VERSION }}
-
-      - name: Set up Python
+      - name: Setup 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 uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
 
       - name: Install dependencies
-        shell: bash
-        run: poetry install --with dev
+        run: pnpm install
 
       - name: Run mypy
         shell: bash
-        run: poetry run mypy llama_index
+        run: uv run mypy llama_index
 
   build:
     needs: [unit-test, type-check]
     runs-on: ubuntu-latest
     defaults:
       run:
-        working-directory: llama-index-server
+        working-directory: python/llama-index-server
     steps:
       - uses: actions/checkout@v4
-      - name: Install Poetry
-        run: pipx install poetry==${{ env.POETRY_VERSION }}
+      - uses: pnpm/action-setup@v3
+
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: ${{ env.PYTHON_VERSION }}
-      - name: Clear python cache
-        shell: bash
-        run: poetry cache clear --all pypi
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
+
+      - name: Install dependencies
+        run: pnpm install && pnpm build
+
       - name: Build package
         shell: bash
-        run: poetry build
-      - name: Test installing built package
+        run: uv build
+
+      - name: Get the absolute wheel file path and save it to the output
         shell: bash
-        run: python -m pip install .
+        id: get_whl_path
+        run: |
+          WHL_FILE=$(readlink -f dist/*.whl)
+          echo "whl_file=$WHL_FILE" >> $GITHUB_OUTPUT
+
       - name: Test import
         shell: bash
-        working-directory: ${{ vars.RUNNER_TEMP }}
-        run: python -c "from llama_index.server import LlamaIndexServer"
+        working-directory: ${{ github.workspace }}
+        env:
+          WHL_FILE: ${{ steps.get_whl_path.outputs.whl_file }}
+        run: |
+          uv run --with $WHL_FILE python -c "from llama_index.server import LlamaIndexServer"
+
+      - name: Check frontend resources is present
+        shell: bash
+        working-directory: ${{ github.workspace }}
+        env:
+          WHL_FILE: ${{ steps.get_whl_path.outputs.whl_file }}
+        run: |
+          uv run --with $WHL_FILE python -c "from llama_index.server.chat_ui import check_ui_resources; check_ui_resources()"
+
       - name: Upload artifact
         uses: actions/upload-artifact@v4
         with:
           name: llama-index-server
-          path: llama-index-server/dist/
+          path: dist/

.gitignore

@@ -6,10 +6,6 @@ node_modules
 .pnpm-store
 .pnp.js
 
-# testing
-coverage
-.coverage
-
 # next.js
 .next/
 out/
@@ -35,31 +31,9 @@ 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

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

.prettierignore

@@ -1,6 +1,18 @@
-apps/docs/i18n
-apps/docs/docs/api
+node_modules/
 pnpm-lock.yaml
 lib/
 dist/
-.docusaurus/
+cache/
+build/
+.next/
+out/
+packages/server/server/
+packages/server/project/
+**/playwright-report/
+**/test-results/
+
+# Python
+python/
+**/*.mypy_cache/**
+**/*.venv/**
+**/*.ruff_cache/**

CLAUDE.md

@@ -0,0 +1,201 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Repository Overview
+
+Create-llama is a monorepo containing CLI tools and server frameworks for building LlamaIndex-powered applications. The repository combines TypeScript/Node.js and Python components in a unified development environment.
+
+## Architecture
+
+### Monorepo Structure
+
+- **`packages/create-llama/`**: Main CLI tool for scaffolding LlamaIndex applications
+- **`packages/server/`**: TypeScript/Next.js server framework (`@llamaindex/server`)
+- **`python/llama-index-server/`**: Python/FastAPI server framework
+- **Root**: Workspace configuration and shared development tools
+
+### Key Technologies
+
+- **Package Manager**: pnpm with workspace configuration
+- **Build Tools**: bunchee (TypeScript), Next.js, hatchling (Python)
+- **Testing**: Playwright for e2e, pytest for Python
+- **Version Management**: changesets for TypeScript packages, manual for Python
+
+## Development Commands
+
+### Root Level (Monorepo)
+
+```bash
+pnpm dev          # Start all packages in development mode
+pnpm build        # Build all packages
+pnpm lint         # ESLint across TypeScript packages
+pnpm format       # Prettier formatting
+pnpm e2e          # Run end-to-end tests
+```
+
+### Create-llama Package
+
+```bash
+cd packages/create-llama
+npm run build     # Build CLI using bash script and ncc
+npm run dev       # Watch mode development
+npm run e2e       # Playwright tests for generated projects
+npm run clean     # Clean build artifacts and template caches
+```
+
+### TypeScript Server Package
+
+```bash
+cd packages/server
+pnpm dev          # Watch mode with bunchee
+pnpm build        # Multi-step build: ESM/CJS + Next.js + static assets
+pnpm clean        # Clean all build outputs
+```
+
+### Python Server Package
+
+```bash
+cd python/llama-index-server
+uv run generate   # Index data files
+fastapi dev       # Start development server with hot reload
+pytest            # Run test suite
+```
+
+## Template System
+
+The CLI uses a sophisticated template system in `packages/create-llama/templates/`:
+
+### Organization
+
+- **`types/`**: Base project structures (streaming, reflex, llamaindexserver)
+- **`components/`**: Reusable components across frameworks
+  - `engines/` - Chat and agent engines
+  - `loaders/` - File, web, database loaders
+  - `providers/` - AI model configurations
+  - `vectordbs/` - Vector database integrations
+  - `use-cases/` - Workflow implementations
+
+### Development Workflow
+
+- Templates support multiple frameworks (Next.js, Express, FastAPI)
+- Component system allows mix-and-match functionality
+- E2E tests validate generated projects work correctly
+
+## Server Framework Architecture
+
+### TypeScript Server (`@llamaindex/server`)
+
+- **Core**: `LlamaIndexServer` class wrapping Next.js with workflow support
+- **Frontend**: React-based chat UI with shadcn/ui components
+- **API**: `/api/chat` endpoint with streaming responses
+- **Build Process**: Complex multi-step build including static assets for Python integration
+
+### Python Server (`llama-index-server`)
+
+- **Core**: `LlamaIndexServer` class extending FastAPI
+- **Architecture**: Workflow factory pattern for stateless request handling
+- **UI Generation**: AI-powered React component generation from Pydantic schemas
+- **Development**: Hot reloading support with dev mode
+
+## Common Patterns
+
+### Workflow Integration
+
+Both server frameworks use factory patterns:
+
+```typescript
+// TypeScript
+const server = new LlamaIndexServer({
+  workflow: (context) => createWorkflow(context)
+});
+
+// Python
+def create_workflow(chat_request: ChatRequest) -> Workflow:
+    return MyWorkflow(chat_request.messages)
+```
+
+### Event System
+
+Structured events for UI communication:
+
+- **UIEvent**: Custom components with Pydantic/Zod schemas
+- **ArtifactEvent**: Code/documents for Canvas panel
+- **SourceNodesEvent**: Document sources with metadata
+- **AgentRunEvent**: Tool usage and progress tracking
+
+### File Handling
+
+- Both servers auto-mount `data/` and `output/` directories
+- LlamaCloud integration for remote file access
+- Static file serving through framework-specific methods
+
+## Testing Strategy
+
+### E2E Testing
+
+- Playwright tests in `packages/create-llama/e2e/`
+- Tests both Python and TypeScript generated projects
+- Validates CLI generation and application functionality
+
+### Unit Testing
+
+- Python: pytest with comprehensive API and service tests
+- TypeScript: Integrated testing through build process
+
+## Build Process
+
+### Create-llama CLI
+
+1. TypeScript compilation with bash script
+2. ncc bundling for standalone executable
+3. Template validation and caching
+
+### Server Package Build
+
+1. **prebuild**: Clean directories
+2. **build**: bunchee compilation to ESM/CJS
+3. **postbuild**: Next.js preparation and static asset generation
+4. **prepare:py-static**: Python integration assets
+
+### Release Process
+
+```bash
+pnpm release     # Build all + publish npm packages + Python release
+```
+
+## Development Environment Setup
+
+### Prerequisites
+
+- Node.js >=16.14.0
+- Python with uv package manager
+- pnpm for package management
+
+### Common Workflow
+
+1. Clone repository and run `pnpm install`
+2. For CLI development: work in `packages/create-llama/`
+3. For server development: choose TypeScript or Python package
+4. Use `pnpm dev` for concurrent development across packages
+5. Run `pnpm e2e` to validate changes with generated projects
+
+## Special Considerations
+
+### Template Development
+
+- Changes to templates require rebuilding CLI
+- E2E tests validate template functionality across frameworks
+- Template caching system speeds up repeated builds
+
+### Cross-package Dependencies
+
+- Server package builds static assets for Python integration
+- Version synchronization between TypeScript and Python packages
+- Shared UI components and styling across implementations
+
+### Performance
+
+- CLI uses caching for template operations
+- Server frameworks support streaming responses
+- Background processing for file operations and LlamaCloud integration

README.md

@@ -25,13 +25,10 @@ to start the development server. You can then visit [http://localhost:3000](http
 ## What you'll get
 
 - A set of pre-configured use cases to get you started, e.g. Agentic RAG, Data Analysis, Report Generation, etc.
-- A Next.js-powered front-end using components from [shadcn/ui](https://ui.shadcn.com/). The app is set up as a chat interface that can answer questions about your data or interact with your agent
-- Your choice of two back-ends:
-  - **Next.js**: if you select this option, you’ll have a full-stack Next.js application that you can deploy to a host like [Vercel](https://vercel.com/) in just a few clicks. This uses [LlamaIndex.TS](https://www.npmjs.com/package/llamaindex), our TypeScript library.
-  - **Python FastAPI**: if you select this option, you’ll get a separate backend powered by the [llama-index Python package](https://pypi.org/project/llama-index/), which you can deploy to a service like [Render](https://render.com/) or [fly.io](https://fly.io/). The separate Next.js front-end will connect to this backend.
-- Each back-end has two endpoints:
-  - One streaming chat endpoint, that allow you to send the state of your chat and receive additional responses
-  - One endpoint to upload private files which can be used in your chat
+- A front-end using components from [shadcn/ui](https://ui.shadcn.com/). The app is set up as a chat interface that can answer questions about your data or interact with your agent
+- Your choice of two frameworks:
+  - **Next.js**: if you select this option, you’ll have a full-stack Next.js application that you can deploy to a host like [Vercel](https://vercel.com/) in just a few clicks. This uses [LlamaIndex.TS](https://www.npmjs.com/package/llamaindex), our TypeScript library with [LlamaIndex Server for TS](https://npmjs.com/package/@llamaindex/server).
+  - **Python FastAPI**: if you select this option, you’ll get full-stack Python application powered by the [llama-index Python package](https://pypi.org/project/llama-index/) and [LlamaIndex Server for Python](https://pypi.org/project/llama-index-server/)
 - The app uses OpenAI by default, so you'll need an OpenAI API key, or you can customize it to use any of the dozens of LLMs we support.
 
 Here's how it looks like:
@@ -40,11 +37,11 @@ https://github.com/user-attachments/assets/d57af1a1-d99b-4e9c-98d9-4cbd1327eff8
 
 ## Using your data
 
-Optionally, you can supply your own data; the app will index it and make use of it, e.g. to answer questions. Your generated app will have a folder called `data` (If you're using Express or Python and generate a frontend, it will be `./backend/data`).
+Optionally, you can supply your own data; the app will index it and make use of it, e.g. to answer questions. Your generated app will have a folder called `data`.
 
-The app will ingest any supported files you put in this directory. Your Next.js and Express apps use LlamaIndex.TS, so they will be able to ingest any PDF, text, CSV, Markdown, Word and HTML files. The Python backend can read even more types, including video and audio files.
+The app will ingest any supported files you put in this directory. Your Next.js apps use LlamaIndex.TS, so they will be able to ingest any PDF, text, CSV, Markdown, Word and HTML files. The Python backend can read even more types, including video and audio files.
 
-Before you can use your data, you need to index it. If you're using the Next.js or Express apps, run:
+Before you can use your data, you need to index it. If you're using the Next.js apps, run:
 
 ```bash
 npm run generate
@@ -55,16 +52,16 @@ 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
-poetry run generate
+uv run generate
 ```
 
 ## Customizing the AI models
 
-The app will default to OpenAI's `gpt-4o-mini` LLM and `text-embedding-3-large` embedding model.
+The app will default to OpenAI's `gpt-4.1` LLM and `text-embedding-3-large` embedding model.
 
-If you want to use different OpenAI models, add the `--ask-models` CLI parameter.
+If you want to use different models, add the `--ask-models` CLI parameter.
 
-You can also replace OpenAI with one of our [dozens of other supported LLMs](https://docs.llamaindex.ai/en/stable/module_guides/models/llms/modules.html).
+You can also replace one of the default models with one of our [dozens of other supported LLMs](https://docs.llamaindex.ai/en/stable/module_guides/models/llms/modules.html).
 
 To do so, you have to manually change the generated code (edit the `settings.ts` file for Typescript projects or the `settings.py` file for Python projects)
 
@@ -90,11 +87,10 @@ Need to install the following packages:
   create-llama@latest
 Ok to proceed? (y) y
 ✔ What is your project named? … my-app
-✔ What app do you want to build? › Agentic RAG
+✔ What use case do you want to build? › Agentic RAG
 ✔ What language do you want to use? › Python (FastAPI)
 ✔ Do you want to use LlamaCloud services? … No / Yes
 ✔ Please provide your LlamaCloud API key (leave blank to skip): …
-✔ Please provide your OpenAI API key (leave blank to skip): …
 ? How would you like to proceed? › - Use arrow-keys. Return to submit.
     Just generate code (~1 sec)
 ❯   Start in VSCode (~1 sec)
@@ -106,28 +102,16 @@ Ok to proceed? (y) y
 You can also pass command line arguments to set up a new project
 non-interactively. For a list of the latest options, call `create-llama --help`.
 
-### Running in pro mode
-
-If you prefer more advanced customization options, you can run `create-llama` in pro mode using the `--pro` flag.
-
-In pro mode, instead of selecting a predefined use case, you'll be prompted to select each technical component of your project. This allows for greater flexibility in customizing your project, including:
-
-- **Vector Store**: Choose from a variety of vector stores for keeping your documents, including MongoDB, Pinecone, Weaviate, Qdrant and Chroma.
-- **Tools**: Choose from a variety of agent tools (functions called by the LLM), such as:
-  - Code Interpreter: Executes Python code in a secure Jupyter notebook environment
-  - Artifact Code Generator: Generates code artifacts that can be run in a sandbox
-  - OpenAPI Action: Facilitates requests to a provided OpenAPI schema
-  - Image Generator: Creates images based on text descriptions
-  - Web Search: Performs web searches to retrieve up-to-date information
-- **Data Sources**: Integrate various data sources into your chat application, including local files, websites, or database-retrieved data.
-- **Backend Options**: Besides using Next.js or FastAPI, you can also select to use Express for a more traditional Node.js application.
-- **Observability**: Choose from a variety of LLM observability tools, including LlamaTrace and Traceloop.
-
-Pro mode is ideal for developers who want fine-grained control over their project's configuration and are comfortable with more technical setup options.
-
 ## LlamaIndex Documentation
 
 - [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)

e2e/python/resolve_dependencies.spec.ts

@@ -1,233 +0,0 @@
-import { expect, test } from "@playwright/test";
-import { exec } from "child_process";
-import fs from "fs";
-import path from "path";
-import util from "util";
-import { TemplateFramework, TemplateVectorDB } from "../../helpers/types";
-import { RunCreateLlamaOptions, createTestDir, runCreateLlama } from "../utils";
-
-const execAsync = util.promisify(exec);
-
-const templateFramework: TemplateFramework = process.env.FRAMEWORK
-  ? (process.env.FRAMEWORK as TemplateFramework)
-  : "fastapi";
-const dataSource: string = process.env.DATASOURCE
-  ? process.env.DATASOURCE
-  : "--example-file";
-
-// TODO: add support for other templates
-
-if (
-  dataSource === "--example-file" // XXX: this test provides its own data source - only trigger it on one data source (usually the CI matrix will trigger multiple data sources)
-) {
-  // vectorDBs, tools, and data source combinations to test
-  const vectorDbs: TemplateVectorDB[] = [
-    "mongo",
-    "pg",
-    "pinecone",
-    "milvus",
-    "astra",
-    "qdrant",
-    "chroma",
-    "weaviate",
-  ];
-
-  const toolOptions = [
-    "wikipedia.WikipediaToolSpec",
-    "google.GoogleSearchToolSpec",
-    "document_generator",
-    "artifact",
-  ];
-
-  const dataSources = [
-    "--example-file",
-    "--web-source https://www.example.com",
-    "--db-source mysql+pymysql://user:pass@localhost:3306/mydb",
-  ];
-
-  const observabilityOptions = ["llamatrace", "traceloop"];
-
-  test.describe("Mypy check", () => {
-    test.describe.configure({ retries: 0 });
-
-    // Test vector databases
-    for (const vectorDb of vectorDbs) {
-      test(`Mypy check for vectorDB: ${vectorDb}`, async () => {
-        const cwd = await createTestDir();
-        const { pyprojectPath } = await createAndCheckLlamaProject({
-          options: {
-            cwd,
-            templateType: "streaming",
-            templateFramework,
-            dataSource: "--example-file",
-            vectorDb,
-            tools: "none",
-            port: 3000,
-            postInstallAction: "none",
-            templateUI: undefined,
-            appType: "--no-frontend",
-            llamaCloudProjectName: undefined,
-            llamaCloudIndexName: undefined,
-            observability: undefined,
-          },
-        });
-
-        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-        if (vectorDb !== "none") {
-          if (vectorDb === "pg") {
-            expect(pyprojectContent).toContain(
-              "llama-index-vector-stores-postgres",
-            );
-          } else {
-            expect(pyprojectContent).toContain(
-              `llama-index-vector-stores-${vectorDb}`,
-            );
-          }
-        }
-      });
-    }
-
-    // Test tools
-    for (const tool of toolOptions) {
-      test(`Mypy check for tool: ${tool}`, async () => {
-        const cwd = await createTestDir();
-        const { pyprojectPath } = await createAndCheckLlamaProject({
-          options: {
-            cwd,
-            templateType: "streaming",
-            templateFramework,
-            dataSource: "--example-file",
-            vectorDb: "none",
-            tools: tool,
-            port: 3000,
-            postInstallAction: "none",
-            templateUI: undefined,
-            appType: "--no-frontend",
-            llamaCloudProjectName: undefined,
-            llamaCloudIndexName: undefined,
-            observability: undefined,
-          },
-        });
-
-        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-        if (tool === "wikipedia.WikipediaToolSpec") {
-          expect(pyprojectContent).toContain("wikipedia");
-        }
-        if (tool === "google.GoogleSearchToolSpec") {
-          expect(pyprojectContent).toContain("google");
-        }
-      });
-    }
-
-    // Test data sources
-    for (const dataSource of dataSources) {
-      const dataSourceType = dataSource.split(" ")[0];
-      test(`Mypy check for data source: ${dataSourceType}`, async () => {
-        const cwd = await createTestDir();
-        const { pyprojectPath } = await createAndCheckLlamaProject({
-          options: {
-            cwd,
-            templateType: "streaming",
-            templateFramework,
-            dataSource,
-            vectorDb: "none",
-            tools: "none",
-            port: 3000,
-            postInstallAction: "none",
-            templateUI: undefined,
-            appType: "--no-frontend",
-            llamaCloudProjectName: undefined,
-            llamaCloudIndexName: undefined,
-            observability: undefined,
-          },
-        });
-
-        const pyprojectContent = fs.readFileSync(pyprojectPath, "utf-8");
-        if (dataSource.includes("--web-source")) {
-          expect(pyprojectContent).toContain("llama-index-readers-web");
-        }
-        if (dataSource.includes("--db-source")) {
-          expect(pyprojectContent).toContain("llama-index-readers-database");
-        }
-      });
-    }
-
-    // Test observability options
-    for (const observability of observabilityOptions) {
-      test(`Mypy check for observability: ${observability}`, async () => {
-        const cwd = await createTestDir();
-
-        const { pyprojectPath } = await createAndCheckLlamaProject({
-          options: {
-            cwd,
-            templateType: "streaming",
-            templateFramework,
-            dataSource: "--example-file",
-            vectorDb: "none",
-            tools: "none",
-            port: 3000,
-            postInstallAction: "none",
-            templateUI: undefined,
-            appType: "--no-frontend",
-            llamaCloudProjectName: undefined,
-            llamaCloudIndexName: undefined,
-            observability,
-          },
-        });
-      });
-    }
-  });
-}
-
-async function createAndCheckLlamaProject({
-  options,
-}: {
-  options: RunCreateLlamaOptions;
-}): Promise<{ pyprojectPath: string; projectPath: string }> {
-  const result = await runCreateLlama(options);
-  const name = result.projectName;
-  const projectPath = path.join(options.cwd, name);
-
-  // Check if the app folder exists
-  expect(fs.existsSync(projectPath)).toBeTruthy();
-
-  // Check if pyproject.toml exists
-  const pyprojectPath = path.join(projectPath, "pyproject.toml");
-  expect(fs.existsSync(pyprojectPath)).toBeTruthy();
-
-  const env = {
-    ...process.env,
-    POETRY_VIRTUALENVS_IN_PROJECT: "true",
-  };
-
-  // Run poetry install
-  try {
-    const { stdout: installStdout, stderr: installStderr } = await execAsync(
-      "poetry install",
-      { cwd: projectPath, env },
-    );
-    console.log("poetry install stdout:", installStdout);
-    console.error("poetry install stderr:", installStderr);
-  } catch (error) {
-    console.error("Error running poetry install:", error);
-    throw error;
-  }
-
-  // Run poetry run mypy
-  try {
-    const { stdout: mypyStdout, stderr: mypyStderr } = await execAsync(
-      "poetry run mypy .",
-      { cwd: projectPath, env },
-    );
-    console.log("poetry run mypy stdout:", mypyStdout);
-    console.error("poetry run mypy stderr:", mypyStderr);
-  } catch (error) {
-    console.error("Error running mypy:", error);
-    throw error;
-  }
-
-  // If we reach this point without throwing an error, the test passes
-  expect(true).toBeTruthy();
-
-  return { pyprojectPath, projectPath };
-}

e2e/shared/reflex_template.spec.ts

@@ -1,64 +0,0 @@
-/* eslint-disable turbo/no-undeclared-env-vars */
-import { expect, test } from "@playwright/test";
-import { ChildProcess } from "child_process";
-import fs from "fs";
-import path from "path";
-import { TemplateFramework, TemplateUseCase } from "../../helpers";
-import { createTestDir, runCreateLlama } from "../utils";
-
-const templateFramework: TemplateFramework = process.env.FRAMEWORK
-  ? (process.env.FRAMEWORK as TemplateFramework)
-  : "fastapi";
-const dataSource: string = process.env.DATASOURCE
-  ? process.env.DATASOURCE
-  : "--example-file";
-const templateUseCases: TemplateUseCase[] = ["extractor", "contract_review"];
-
-// The reflex template currently only works with FastAPI and files (and not on Windows)
-if (
-  process.platform !== "win32" &&
-  templateFramework === "fastapi" &&
-  dataSource === "--example-file"
-) {
-  for (const useCase of templateUseCases) {
-    test.describe(`Test reflex template ${useCase} ${templateFramework} ${dataSource}`, async () => {
-      let appPort: number;
-      let name: string;
-      let appProcess: ChildProcess;
-      let cwd: string;
-
-      // Create reflex app
-      test.beforeAll(async () => {
-        cwd = await createTestDir();
-        appPort = Math.floor(Math.random() * 10000) + 10000;
-        const result = await runCreateLlama({
-          cwd,
-          templateType: "reflex",
-          templateFramework: "fastapi",
-          dataSource: "--example-file",
-          vectorDb: "none",
-          port: appPort,
-          postInstallAction: "runApp",
-          useCase,
-        });
-        name = result.projectName;
-        appProcess = result.appProcess;
-      });
-
-      test.afterAll(async () => {
-        appProcess.kill();
-      });
-
-      test("App folder should exist", async () => {
-        const dirExists = fs.existsSync(path.join(cwd, name));
-        expect(dirExists).toBeTruthy();
-      });
-      test("Frontend should have a title", async ({ page }) => {
-        await page.goto(`http://localhost:${appPort}`);
-        await expect(page.getByText("Built by LlamaIndex")).toBeVisible({
-          timeout: 2000 * 60,
-        });
-      });
-    });
-  }
-}

e2e/shared/streaming_template.spec.ts

@@ -1,128 +0,0 @@
-/* eslint-disable turbo/no-undeclared-env-vars */
-import { expect, test } from "@playwright/test";
-import { ChildProcess } from "child_process";
-import fs from "fs";
-import path from "path";
-import type {
-  TemplateFramework,
-  TemplatePostInstallAction,
-  TemplateUI,
-} from "../../helpers";
-import { createTestDir, runCreateLlama, type AppType } from "../utils";
-
-const templateFramework: TemplateFramework = process.env.FRAMEWORK
-  ? (process.env.FRAMEWORK as TemplateFramework)
-  : "fastapi";
-const dataSource: string = process.env.DATASOURCE
-  ? process.env.DATASOURCE
-  : "--example-file";
-const templateUI: TemplateUI = "shadcn";
-const templatePostInstallAction: TemplatePostInstallAction = "runApp";
-
-const llamaCloudProjectName = "create-llama";
-const llamaCloudIndexName = "e2e-test";
-
-const appType: AppType = templateFramework === "fastapi" ? "--frontend" : "";
-const userMessage =
-  dataSource !== "--no-files" ? "Physical standard for letters" : "Hello";
-
-test.describe(`Test streaming template ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
-  const isNode18 = process.version.startsWith("v18");
-  const isLlamaCloud = dataSource === "--llamacloud";
-  // llamacloud is using File API which is not supported on node 18
-  if (isNode18 && isLlamaCloud) {
-    test.skip(true, "Skipping tests for Node 18 and LlamaCloud data source");
-  }
-
-  let port: number;
-  let cwd: string;
-  let name: string;
-  let appProcess: ChildProcess;
-  // Only test without using vector db for now
-  const vectorDb = "none";
-
-  test.beforeAll(async () => {
-    port = Math.floor(Math.random() * 10000) + 10000;
-    cwd = await createTestDir();
-    const result = await runCreateLlama({
-      cwd,
-      templateType: "streaming",
-      templateFramework,
-      dataSource,
-      vectorDb,
-      port,
-      postInstallAction: templatePostInstallAction,
-      templateUI,
-      appType,
-      llamaCloudProjectName,
-      llamaCloudIndexName,
-    });
-    name = result.projectName;
-    appProcess = result.appProcess;
-  });
-
-  test("App folder should exist", async () => {
-    const dirExists = fs.existsSync(path.join(cwd, name));
-    expect(dirExists).toBeTruthy();
-  });
-
-  test("Frontend should have a title", async ({ page }) => {
-    test.skip(
-      templatePostInstallAction !== "runApp" || templateFramework === "express",
-    );
-    await page.goto(`http://localhost:${port}`);
-    await expect(page.getByText("Built by LlamaIndex")).toBeVisible();
-  });
-
-  test("Frontend should be able to submit a message and receive a response", async ({
-    page,
-  }) => {
-    test.skip(
-      templatePostInstallAction !== "runApp" || templateFramework === "express",
-    );
-    await page.goto(`http://localhost:${port}`);
-    await page.fill("form textarea", userMessage);
-    const [response] = await Promise.all([
-      page.waitForResponse(
-        (res) => {
-          return res.url().includes("/api/chat") && res.status() === 200;
-        },
-        {
-          timeout: 1000 * 60,
-        },
-      ),
-      page.click("form button[type=submit]"),
-    ]);
-    const text = await response.text();
-    console.log("AI response when submitting message: ", text);
-    expect(response.ok()).toBeTruthy();
-  });
-
-  test("Backend frameworks should response when calling non-streaming chat API", async ({
-    request,
-  }) => {
-    test.skip(templatePostInstallAction !== "runApp");
-    test.skip(templateFramework === "nextjs");
-    const response = await request.post(
-      `http://localhost:${port}/api/chat/request`,
-      {
-        data: {
-          messages: [
-            {
-              role: "user",
-              content: userMessage,
-            },
-          ],
-        },
-      },
-    );
-    const text = await response.text();
-    console.log("AI response when calling API: ", text);
-    expect(response.ok()).toBeTruthy();
-  });
-
-  // clean processes
-  test.afterAll(async () => {
-    appProcess?.kill();
-  });
-});

e2e/typescript/resolve_dependencies.spec.ts

@@ -1,105 +0,0 @@
-import { expect, test } from "@playwright/test";
-import { exec } from "child_process";
-import fs from "fs";
-import path from "path";
-import util from "util";
-import { TemplateFramework, TemplateVectorDB } from "../../helpers/types";
-import { createTestDir, runCreateLlama } from "../utils";
-
-const execAsync = util.promisify(exec);
-
-const templateFramework: TemplateFramework = process.env.FRAMEWORK
-  ? (process.env.FRAMEWORK as TemplateFramework)
-  : "nextjs";
-const dataSource: string = process.env.DATASOURCE
-  ? process.env.DATASOURCE
-  : "--example-file";
-
-// vectorDBs combinations to test
-const vectorDbs: TemplateVectorDB[] = [
-  "mongo",
-  "pg",
-  "qdrant",
-  "pinecone",
-  "milvus",
-  "astra",
-  "chroma",
-  "llamacloud",
-  "weaviate",
-];
-
-test.describe("Test resolve TS dependencies", () => {
-  // Test vector DBs without LlamaParse
-  for (const vectorDb of vectorDbs) {
-    const optionDescription = `vectorDb: ${vectorDb}, dataSource: ${dataSource}`;
-
-    test(`Vector DB test - ${optionDescription}`, async () => {
-      await runTest(vectorDb, false);
-    });
-  }
-
-  // Test LlamaParse with vectorDB 'none'
-  test(`LlamaParse test - vectorDb: none, dataSource: ${dataSource}, llamaParse: true`, async () => {
-    await runTest("none", true);
-  });
-
-  async function runTest(
-    vectorDb: TemplateVectorDB | "none",
-    useLlamaParse: boolean,
-  ) {
-    const cwd = await createTestDir();
-
-    const result = await runCreateLlama({
-      cwd: cwd,
-      templateType: "streaming",
-      templateFramework: templateFramework,
-      dataSource: dataSource,
-      vectorDb: vectorDb,
-      port: 3000,
-      postInstallAction: "none",
-      templateUI: undefined,
-      appType: templateFramework === "nextjs" ? "" : "--no-frontend",
-      llamaCloudProjectName: undefined,
-      llamaCloudIndexName: undefined,
-      tools: undefined,
-      useLlamaParse: useLlamaParse,
-    });
-    const name = result.projectName;
-
-    // Check if the app folder exists
-    const appDir = path.join(cwd, name);
-    const dirExists = fs.existsSync(appDir);
-    expect(dirExists).toBeTruthy();
-
-    // Install dependencies using pnpm
-    try {
-      const { stderr: installStderr } = await execAsync(
-        "pnpm install --prefer-offline",
-        {
-          cwd: appDir,
-        },
-      );
-    } catch (error) {
-      console.error("Error installing dependencies:", error);
-      throw error;
-    }
-
-    // Run tsc type check and capture the output
-    try {
-      const { stdout, stderr } = await execAsync(
-        "pnpm exec tsc -b --diagnostics",
-        {
-          cwd: appDir,
-        },
-      );
-      // Check if there's any error output
-      expect(stderr).toBeFalsy();
-
-      // Log the stdout for debugging purposes
-      console.log("TypeScript type-check output:", stdout);
-    } catch (error) {
-      console.error("Error running tsc:", error);
-      throw error;
-    }
-  }
-});

eslint.config.mjs

@@ -0,0 +1,65 @@
+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/**",
+      "packages/server/server/**",
+      "packages/server/project/**",
+      "packages/server/bin/**",
+    ],
+  },
+);

helpers/constant.ts

@@ -1,6 +0,0 @@
-export const COMMUNITY_OWNER = "run-llama";
-export const COMMUNITY_REPO = "create_llama_projects";
-export const LLAMA_PACK_OWNER = "run-llama";
-export const LLAMA_PACK_REPO = "llama_index";
-export const LLAMA_PACK_FOLDER = "llama-index-packs";
-export const LLAMA_PACK_FOLDER_PATH = `${LLAMA_PACK_OWNER}/${LLAMA_PACK_REPO}/main/${LLAMA_PACK_FOLDER}`;

helpers/datasources.ts

@@ -1,142 +0,0 @@
-import fs from "fs/promises";
-import path from "path";
-import yaml, { Document } from "yaml";
-import { templatesDir } from "./dir";
-import { DbSourceConfig, TemplateDataSource, WebSourceConfig } from "./types";
-
-export const EXAMPLE_FILE: TemplateDataSource = {
-  type: "file",
-  config: {
-    path: path.join(templatesDir, "components", "data", "101.pdf"),
-  },
-};
-
-export const EXAMPLE_10K_SEC_FILES: TemplateDataSource[] = [
-  {
-    type: "file",
-    config: {
-      url: new URL(
-        "https://s2.q4cdn.com/470004039/files/doc_earnings/2023/q4/filing/_10-K-Q4-2023-As-Filed.pdf",
-      ),
-      filename: "apple_10k_report.pdf",
-    },
-  },
-  {
-    type: "file",
-    config: {
-      url: new URL(
-        "https://ir.tesla.com/_flysystem/s3/sec/000162828024002390/tsla-20231231-gen.pdf",
-      ),
-      filename: "tesla_10k_report.pdf",
-    },
-  },
-];
-
-export const EXAMPLE_GDPR: TemplateDataSource = {
-  type: "file",
-  config: {
-    url: new URL(
-      "https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32016R0679",
-    ),
-    filename: "gdpr.pdf",
-  },
-};
-
-export const AI_REPORTS: TemplateDataSource = {
-  type: "file",
-  config: {
-    url: new URL(
-      "https://www.europarl.europa.eu/RegData/etudes/ATAG/2024/760392/EPRS_ATA(2024)760392_EN.pdf",
-    ),
-    filename: "EPRS_ATA_2024_760392_EN.pdf",
-  },
-};
-
-export function getDataSources(
-  files?: string,
-  exampleFile?: boolean,
-): TemplateDataSource[] | undefined {
-  let dataSources: TemplateDataSource[] | undefined = undefined;
-  if (files) {
-    // If user specified files option, then the program should use context engine
-    dataSources = files.split(",").map((filePath) => ({
-      type: "file",
-      config: {
-        path: filePath,
-      },
-    }));
-  }
-  if (exampleFile) {
-    dataSources = [...(dataSources ? dataSources : []), EXAMPLE_FILE];
-  }
-  return dataSources;
-}
-
-export async function writeLoadersConfig(
-  root: string,
-  dataSources: TemplateDataSource[],
-  useLlamaParse?: boolean,
-) {
-  const loaderConfig: Record<string, any> = {};
-
-  // Always set file loader config
-  loaderConfig.file = createFileLoaderConfig(useLlamaParse);
-
-  if (dataSources.some((ds) => ds.type === "web")) {
-    loaderConfig.web = createWebLoaderConfig(dataSources);
-  }
-
-  const dbLoaders = dataSources.filter((ds) => ds.type === "db");
-  if (dbLoaders.length > 0) {
-    loaderConfig.db = createDbLoaderConfig(dbLoaders);
-  }
-
-  // Create a new Document with the loaderConfig
-  const yamlDoc = new Document(loaderConfig);
-
-  // Write loaders config
-  const loaderConfigPath = path.join(root, "config", "loaders.yaml");
-  await fs.mkdir(path.join(root, "config"), { recursive: true });
-  await fs.writeFile(loaderConfigPath, yaml.stringify(yamlDoc));
-}
-
-function createWebLoaderConfig(dataSources: TemplateDataSource[]): any {
-  const webLoaderConfig: Record<string, any> = {};
-
-  // Create config for browser driver arguments
-  webLoaderConfig.driver_arguments = [
-    "--no-sandbox",
-    "--disable-dev-shm-usage",
-  ];
-
-  // Create config for urls
-  const urlConfigs = dataSources
-    .filter((ds) => ds.type === "web")
-    .map((ds) => {
-      const dsConfig = ds.config as WebSourceConfig;
-      return {
-        base_url: dsConfig.baseUrl,
-        prefix: dsConfig.prefix,
-        depth: dsConfig.depth,
-      };
-    });
-  webLoaderConfig.urls = urlConfigs;
-
-  return webLoaderConfig;
-}
-
-function createFileLoaderConfig(useLlamaParse?: boolean): any {
-  return {
-    use_llama_parse: useLlamaParse,
-  };
-}
-
-function createDbLoaderConfig(dbLoaders: TemplateDataSource[]): any {
-  return dbLoaders.map((ds) => {
-    const dsConfig = ds.config as DbSourceConfig;
-    return {
-      uri: dsConfig.uri,
-      queries: [dsConfig.queries],
-    };
-  });
-}

helpers/llama-pack.ts

@@ -1,148 +0,0 @@
-import fs from "fs/promises";
-import got from "got";
-import path from "path";
-import { parse } from "smol-toml";
-import {
-  LLAMA_PACK_FOLDER,
-  LLAMA_PACK_FOLDER_PATH,
-  LLAMA_PACK_OWNER,
-  LLAMA_PACK_REPO,
-} from "./constant";
-import { copy } from "./copy";
-import { templatesDir } from "./dir";
-import { addDependencies, installPythonDependencies } from "./python";
-import { getRepoRawContent } from "./repo";
-import { InstallTemplateArgs } from "./types";
-
-const getLlamaPackFolderSHA = async () => {
-  const url = `https://api.github.com/repos/${LLAMA_PACK_OWNER}/${LLAMA_PACK_REPO}/contents`;
-  const response = await got(url, {
-    responseType: "json",
-  });
-  const data = response.body as any[];
-  const llamaPackFolder = data.find((item) => item.name === LLAMA_PACK_FOLDER);
-  return llamaPackFolder.sha;
-};
-
-const getLLamaPackFolderTree = async (
-  sha: string,
-): Promise<
-  Array<{
-    path: string;
-  }>
-> => {
-  const url = `https://api.github.com/repos/${LLAMA_PACK_OWNER}/${LLAMA_PACK_REPO}/git/trees/${sha}?recursive=1`;
-  const response = await got(url, {
-    responseType: "json",
-  });
-  return (response.body as any).tree;
-};
-
-export async function getAvailableLlamapackOptions(): Promise<
-  {
-    name: string;
-    folderPath: string;
-  }[]
-> {
-  const EXAMPLE_RELATIVE_PATH = "/examples/example.py";
-  const PACK_FOLDER_SUBFIX = "llama-index-packs";
-
-  const llamaPackFolderSHA = await getLlamaPackFolderSHA();
-  const llamaPackTree = await getLLamaPackFolderTree(llamaPackFolderSHA);
-
-  // Return options that have example files
-  const exampleFiles = llamaPackTree.filter((item) =>
-    item.path.endsWith(EXAMPLE_RELATIVE_PATH),
-  );
-  const options = exampleFiles.map((file) => {
-    const packFolder = file.path.substring(
-      0,
-      file.path.indexOf(EXAMPLE_RELATIVE_PATH),
-    );
-    const packName = packFolder.substring(PACK_FOLDER_SUBFIX.length + 1);
-    return {
-      name: packName,
-      folderPath: packFolder,
-    };
-  });
-  return options;
-}
-
-const copyLlamapackEmptyProject = async ({
-  root,
-}: Pick<InstallTemplateArgs, "root">) => {
-  const templatePath = path.join(
-    templatesDir,
-    "components/sample-projects/llamapack",
-  );
-  await copy("**", root, {
-    parents: true,
-    cwd: templatePath,
-  });
-};
-
-const copyData = async ({
-  root,
-}: Pick<InstallTemplateArgs, "root" | "llamapack">) => {
-  const dataPath = path.join(templatesDir, "components/data");
-  await copy("**", path.join(root, "data"), {
-    parents: true,
-    cwd: dataPath,
-  });
-};
-
-const installLlamapackExample = async ({
-  root,
-  llamapack,
-}: Pick<InstallTemplateArgs, "root" | "llamapack">) => {
-  const exampleFileName = "example.py";
-  const readmeFileName = "README.md";
-  const projectTomlFileName = "pyproject.toml";
-  const exampleFilePath = `${LLAMA_PACK_FOLDER_PATH}/${llamapack}/examples/${exampleFileName}`;
-  const readmeFilePath = `${LLAMA_PACK_FOLDER_PATH}/${llamapack}/${readmeFileName}`;
-  const projectTomlFilePath = `${LLAMA_PACK_FOLDER_PATH}/${llamapack}/${projectTomlFileName}`;
-
-  // Download example.py from llamapack and save to root
-  const exampleContent = await getRepoRawContent(exampleFilePath);
-  await fs.writeFile(path.join(root, exampleFileName), exampleContent);
-
-  // Download README.md from llamapack and combine with README-template.md,
-  // save to root and then delete template file
-  const readmeContent = await getRepoRawContent(readmeFilePath);
-  const readmeTemplateContent = await fs.readFile(
-    path.join(root, "README-template.md"),
-    "utf-8",
-  );
-  await fs.writeFile(
-    path.join(root, readmeFileName),
-    `${readmeContent}\n${readmeTemplateContent}`,
-  );
-  await fs.unlink(path.join(root, "README-template.md"));
-
-  // Download pyproject.toml from llamapack, parse it to get package name and version,
-  // then add it as a dependency to current toml file in the project
-  const projectTomlContent = await getRepoRawContent(projectTomlFilePath);
-  const fileParsed = parse(projectTomlContent) as any;
-  const packageName = fileParsed.tool.poetry.name;
-  const packageVersion = fileParsed.tool.poetry.version;
-  await addDependencies(root, [
-    {
-      name: packageName,
-      version: packageVersion,
-    },
-  ]);
-};
-
-export const installLlamapackProject = async ({
-  root,
-  llamapack,
-  postInstallAction,
-}: Pick<InstallTemplateArgs, "root" | "llamapack" | "postInstallAction">) => {
-  console.log("\nInstalling Llamapack project:", llamapack!);
-  await copyLlamapackEmptyProject({ root });
-  await copyData({ root });
-  await installLlamapackExample({ root, llamapack });
-  if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
-    installPythonDependencies({ noRoot: true });
-  }
-};

helpers/poetry.ts

@@ -1,36 +0,0 @@
-/* eslint-disable import/no-extraneous-dependencies */
-import { execSync } from "child_process";
-import fs from "fs";
-
-export function isPoetryAvailable(): boolean {
-  try {
-    execSync("poetry --version", { stdio: "ignore" });
-    return true;
-  } catch (_) {}
-  return false;
-}
-
-export function tryPoetryInstall(noRoot: boolean): boolean {
-  try {
-    execSync(`poetry install${noRoot ? " --no-root" : ""}`, {
-      stdio: "inherit",
-    });
-    return true;
-  } catch (_) {}
-  return false;
-}
-
-export function tryPoetryRun(command: string): boolean {
-  try {
-    execSync(`poetry run ${command}`, { stdio: "inherit" });
-    return true;
-  } catch (_) {}
-  return false;
-}
-
-export function isHavingPoetryLockFile(): boolean {
-  try {
-    return fs.existsSync("poetry.lock");
-  } catch (_) {}
-  return false;
-}

helpers/providers/index.ts

@@ -1,94 +0,0 @@
-import prompts from "prompts";
-import { questionHandlers } from "../../questions/utils";
-import { ModelConfig, ModelProvider, TemplateFramework } from "../types";
-import { askAnthropicQuestions } from "./anthropic";
-import { askAzureQuestions } from "./azure";
-import { askGeminiQuestions } from "./gemini";
-import { askGroqQuestions } from "./groq";
-import { askHuggingfaceQuestions } from "./huggingface";
-import { askLLMHubQuestions } from "./llmhub";
-import { askMistralQuestions } from "./mistral";
-import { askOllamaQuestions } from "./ollama";
-import { askOpenAIQuestions } from "./openai";
-
-const DEFAULT_MODEL_PROVIDER = "openai";
-
-export type ModelConfigQuestionsParams = {
-  openAiKey?: string;
-  askModels: boolean;
-  framework?: TemplateFramework;
-};
-
-export type ModelConfigParams = Omit<ModelConfig, "provider">;
-
-export async function askModelConfig({
-  askModels,
-  openAiKey,
-  framework,
-}: ModelConfigQuestionsParams): Promise<ModelConfig> {
-  let modelProvider: ModelProvider = DEFAULT_MODEL_PROVIDER;
-  if (askModels) {
-    let choices = [
-      { title: "OpenAI", value: "openai" },
-      { title: "Groq", value: "groq" },
-      { title: "Ollama", value: "ollama" },
-      { title: "Anthropic", value: "anthropic" },
-      { title: "Gemini", value: "gemini" },
-      { title: "Mistral", value: "mistral" },
-      { title: "AzureOpenAI", value: "azure-openai" },
-    ];
-
-    if (framework === "fastapi") {
-      choices.push({ title: "T-Systems", value: "t-systems" });
-      choices.push({ title: "Huggingface", value: "huggingface" });
-    }
-    const { provider } = await prompts(
-      {
-        type: "select",
-        name: "provider",
-        message: "Which model provider would you like to use",
-        choices: choices,
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    modelProvider = provider;
-  }
-
-  let modelConfig: ModelConfigParams;
-  switch (modelProvider) {
-    case "ollama":
-      modelConfig = await askOllamaQuestions({ askModels });
-      break;
-    case "groq":
-      modelConfig = await askGroqQuestions({ askModels });
-      break;
-    case "anthropic":
-      modelConfig = await askAnthropicQuestions({ askModels });
-      break;
-    case "gemini":
-      modelConfig = await askGeminiQuestions({ askModels });
-      break;
-    case "mistral":
-      modelConfig = await askMistralQuestions({ askModels });
-      break;
-    case "azure-openai":
-      modelConfig = await askAzureQuestions({ askModels });
-      break;
-    case "t-systems":
-      modelConfig = await askLLMHubQuestions({ askModels });
-      break;
-    case "huggingface":
-      modelConfig = await askHuggingfaceQuestions({ askModels });
-      break;
-    default:
-      modelConfig = await askOpenAIQuestions({
-        openAiKey,
-        askModels,
-      });
-  }
-  return {
-    ...modelConfig,
-    provider: modelProvider,
-  };
-}

helpers/python.ts

@@ -1,645 +0,0 @@
-import fs from "fs/promises";
-import path from "path";
-import { cyan, red } from "picocolors";
-import { parse, stringify } from "smol-toml";
-import terminalLink from "terminal-link";
-
-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";
-
-interface Dependency {
-  name: string;
-  version?: string;
-  extras?: string[];
-  constraints?: Record<string, string>;
-}
-
-const getAdditionalDependencies = (
-  modelConfig: ModelConfig,
-  vectorDb?: TemplateVectorDB,
-  dataSources?: TemplateDataSource[],
-  tools?: Tool[],
-  templateType?: TemplateType,
-  observability?: TemplateObservability,
-) => {
-  const dependencies: Dependency[] = [];
-
-  // Add vector db dependencies
-  switch (vectorDb) {
-    case "mongo": {
-      dependencies.push({
-        name: "llama-index-vector-stores-mongodb",
-        version: "^0.6.0",
-      });
-      break;
-    }
-    case "pg": {
-      dependencies.push({
-        name: "llama-index-vector-stores-postgres",
-        version: "^0.3.2",
-      });
-      break;
-    }
-    case "pinecone": {
-      dependencies.push({
-        name: "llama-index-vector-stores-pinecone",
-        version: "^0.4.1",
-        constraints: {
-          python: ">=3.11,<3.13",
-        },
-      });
-      break;
-    }
-    case "milvus": {
-      dependencies.push({
-        name: "llama-index-vector-stores-milvus",
-        version: "^0.3.0",
-      });
-      dependencies.push({
-        name: "pymilvus",
-        version: "2.4.4",
-      });
-      break;
-    }
-    case "astra": {
-      dependencies.push({
-        name: "llama-index-vector-stores-astra-db",
-        version: "^0.4.0",
-      });
-      break;
-    }
-    case "qdrant": {
-      dependencies.push({
-        name: "llama-index-vector-stores-qdrant",
-        version: "^0.4.0",
-        constraints: {
-          python: ">=3.11,<3.13",
-        },
-      });
-      break;
-    }
-    case "chroma": {
-      dependencies.push({
-        name: "llama-index-vector-stores-chroma",
-        version: "^0.4.0",
-      });
-      break;
-    }
-    case "weaviate": {
-      dependencies.push({
-        name: "llama-index-vector-stores-weaviate",
-        version: "^1.2.3",
-      });
-      break;
-    }
-    case "llamacloud":
-      dependencies.push({
-        name: "llama-index-indices-managed-llama-cloud",
-        version: "0.6.3",
-      });
-      break;
-  }
-
-  // Add data source dependencies
-  if (dataSources) {
-    for (const ds of dataSources) {
-      const dsType = ds?.type;
-      switch (dsType) {
-        case "file":
-          dependencies.push({
-            name: "docx2txt",
-            version: "^0.8",
-          });
-          break;
-        case "web":
-          dependencies.push({
-            name: "llama-index-readers-web",
-            version: "^0.3.0",
-          });
-          break;
-        case "db":
-          dependencies.push({
-            name: "llama-index-readers-database",
-            version: "^0.3.0",
-          });
-          dependencies.push({
-            name: "pymysql",
-            version: "^1.1.0",
-            extras: ["rsa"],
-          });
-          dependencies.push({
-            name: "psycopg2-binary",
-            version: "^2.9.9",
-          });
-          break;
-      }
-    }
-  }
-
-  // Add tools dependencies
-  console.log("Adding tools dependencies");
-  tools?.forEach((tool) => {
-    tool.dependencies?.forEach((dep) => {
-      dependencies.push(dep);
-    });
-  });
-
-  switch (modelConfig.provider) {
-    case "ollama":
-      dependencies.push({
-        name: "llama-index-llms-ollama",
-        version: "0.3.0",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-ollama",
-        version: "0.3.0",
-      });
-      break;
-    case "openai":
-      if (templateType !== "multiagent") {
-        dependencies.push({
-          name: "llama-index-llms-openai",
-          version: "^0.3.2",
-        });
-        dependencies.push({
-          name: "llama-index-embeddings-openai",
-          version: "^0.3.1",
-        });
-        dependencies.push({
-          name: "llama-index-agent-openai",
-          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: "0.2.0",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-fastembed",
-        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: "0.3.0",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-fastembed",
-        version: "^0.2.0",
-      });
-      break;
-    case "gemini":
-      dependencies.push({
-        name: "llama-index-llms-gemini",
-        version: "0.3.4",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-gemini",
-        version: "^0.2.0",
-      });
-      break;
-    case "mistral":
-      dependencies.push({
-        name: "llama-index-llms-mistralai",
-        version: "0.2.1",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-mistralai",
-        version: "0.2.0",
-      });
-      break;
-    case "azure-openai":
-      dependencies.push({
-        name: "llama-index-llms-azure-openai",
-        version: "0.2.0",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-azure-openai",
-        version: "0.2.4",
-      });
-      break;
-    case "huggingface":
-      dependencies.push({
-        name: "llama-index-llms-huggingface",
-        version: "^0.3.5",
-      });
-      dependencies.push({
-        name: "llama-index-embeddings-huggingface",
-        version: "^0.3.1",
-      });
-      dependencies.push({
-        name: "optimum",
-        version: "^1.23.3",
-        extras: ["onnxruntime"],
-      });
-      break;
-    case "t-systems":
-      dependencies.push({
-        name: "llama-index-agent-openai",
-        version: "0.3.0",
-      });
-      dependencies.push({
-        name: "llama-index-llms-openai-like",
-        version: "0.2.0",
-      });
-      break;
-  }
-
-  if (observability && observability !== "none") {
-    if (observability === "traceloop") {
-      dependencies.push({
-        name: "traceloop-sdk",
-        version: "^0.15.11",
-      });
-    }
-    if (observability === "llamatrace") {
-      dependencies.push({
-        name: "llama-index-callbacks-arize-phoenix",
-        version: "^0.3.0",
-      });
-    }
-  }
-
-  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 };
-    }
-
-    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;
-    }
-  }
-};
-
-const copyRouterCode = async (root: string, tools: Tool[]) => {
-  // Copy sandbox router if the artifact tool is selected
-  if (tools?.some((t) => t.name === "artifact")) {
-    await copy("sandbox.py", path.join(root, "app", "api", "routers"), {
-      parents: true,
-      cwd: path.join(templatesDir, "components", "routers", "python"),
-      rename: assetRelocator,
-    });
-  }
-};
-
-export const addDependencies = async (
-  projectDir: string,
-  dependencies: Dependency[],
-) => {
-  if (dependencies.length === 0) return;
-
-  const FILENAME = "pyproject.toml";
-  try {
-    // Parse toml file
-    const file = path.join(projectDir, FILENAME);
-    const fileContent = await fs.readFile(file, "utf8");
-    const fileParsed = parse(fileContent);
-
-    // 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);
-
-    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`,
-      error,
-    );
-  }
-};
-
-export const installPythonDependencies = (
-  { noRoot }: { noRoot: boolean } = { noRoot: false },
-) => {
-  if (isPoetryAvailable()) {
-    console.log(
-      `Installing python dependencies using poetry. This may take a while...`,
-    );
-    const installSuccessful = tryPoetryInstall(noRoot);
-    if (!installSuccessful) {
-      console.error(
-        red(
-          "Installing dependencies using poetry failed. Please check error log above and try running create-llama again.",
-        ),
-      );
-      process.exit(1);
-    }
-  } else {
-    console.error(
-      red(
-        `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 ({
-  root,
-  template,
-  vectorDb,
-  dataSources,
-  tools,
-  useCase,
-  observability,
-}: Pick<
-  InstallTemplateArgs,
-  | "root"
-  | "template"
-  | "vectorDb"
-  | "dataSources"
-  | "tools"
-  | "useCase"
-  | "observability"
->) => {
-  const compPath = path.join(templatesDir, "components");
-  const enginePath = path.join(root, "app", "engine");
-
-  // Copy selected vector DB
-  await copy("**", enginePath, {
-    parents: true,
-    cwd: path.join(compPath, "vectordbs", "python", vectorDb ?? "none"),
-  });
-
-  if (vectorDb !== "llamacloud") {
-    // Copy all loaders to enginePath
-    // Not needed for LlamaCloud as it has its own loaders
-    const loaderPath = path.join(enginePath, "loaders");
-    await copy("**", loaderPath, {
-      parents: true,
-      cwd: path.join(compPath, "loaders", "python"),
-    });
-  }
-
-  // Copy settings.py to app
-  await copy("**", path.join(root, "app"), {
-    cwd: path.join(compPath, "settings", "python"),
-  });
-
-  // Copy services
-  if (template == "streaming" || template == "multiagent") {
-    await copy("**", path.join(root, "app", "api", "services"), {
-      cwd: path.join(compPath, "services", "python"),
-    });
-  }
-  // Copy engine code
-  if (template === "streaming" || template === "multiagent") {
-    // Select and copy engine code based on data sources and tools
-    let engine;
-    // Multiagent always uses agent engine
-    if (template === "multiagent") {
-      engine = "agent";
-    } else {
-      // For streaming, use chat engine by default
-      // Unless tools are selected, in which case use agent engine
-      if (dataSources.length > 0 && (!tools || tools.length === 0)) {
-        console.log(
-          "\nNo tools selected - use optimized context chat engine\n",
-        );
-        engine = "chat";
-      } else {
-        engine = "agent";
-      }
-    }
-
-    // Copy engine code
-    await copy("**", enginePath, {
-      parents: true,
-      cwd: path.join(compPath, "engines", "python", engine),
-    });
-
-    // Copy router code
-    await copyRouterCode(root, tools ?? []);
-  }
-
-  // Copy multiagents overrides
-  if (template === "multiagent") {
-    await copy("**", path.join(root), {
-      cwd: path.join(compPath, "multiagent", "python"),
-    });
-  }
-
-  if (template === "multiagent" || template === "reflex") {
-    if (useCase) {
-      const sourcePath =
-        template === "multiagent"
-          ? path.join(compPath, "agents", "python", useCase)
-          : path.join(compPath, "reflex", useCase);
-
-      await copy("**", path.join(root), {
-        parents: true,
-        cwd: sourcePath,
-        rename: assetRelocator,
-      });
-    } else {
-      console.log(
-        red(
-          `There is no use case selected for ${template} template. Please pick a use case to use via --use-case flag.`,
-        ),
-      );
-      process.exit(1);
-    }
-  }
-
-  if (observability && observability !== "none") {
-    const templateObservabilityPath = path.join(
-      templatesDir,
-      "components",
-      "observability",
-      "python",
-      observability,
-    );
-    await copy("**", path.join(root, "app"), {
-      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("workflow.py", path.join(root, "app"), {
-    parents: true,
-    cwd: path.join(templatesDir, "components", "workflows", "python", 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);
-
-  if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
-    installPythonDependencies();
-  }
-};

helpers/repo.ts

@@ -1,134 +0,0 @@
-import { createWriteStream, promises } from "fs";
-import got from "got";
-import { tmpdir } from "os";
-import { join } from "path";
-import { Stream } from "stream";
-import tar from "tar";
-import { promisify } from "util";
-import { makeDir } from "./make-dir";
-import { CommunityProjectConfig } from "./types";
-
-export type RepoInfo = {
-  username: string;
-  name: string;
-  branch: string;
-  filePath: string;
-};
-
-const pipeline = promisify(Stream.pipeline);
-
-async function downloadTar(url: string) {
-  const tempFile = join(tmpdir(), `next.js-cna-example.temp-${Date.now()}`);
-  await pipeline(got.stream(url), createWriteStream(tempFile));
-  return tempFile;
-}
-
-export async function downloadAndExtractRepo(
-  root: string,
-  { username, name, branch, filePath }: RepoInfo,
-) {
-  await makeDir(root);
-
-  const tempFile = await downloadTar(
-    `https://codeload.github.com/${username}/${name}/tar.gz/${branch}`,
-  );
-
-  await tar.x({
-    file: tempFile,
-    cwd: root,
-    strip: filePath ? filePath.split("/").length + 1 : 1,
-    filter: (p) =>
-      p.startsWith(
-        `${name}-${branch.replace(/\//g, "-")}${
-          filePath ? `/${filePath}/` : "/"
-        }`,
-      ),
-  });
-
-  await promises.unlink(tempFile);
-}
-
-const getRepoInfo = async (owner: string, repo: string) => {
-  const repoInfoRes = await got(
-    `https://api.github.com/repos/${owner}/${repo}`,
-    {
-      responseType: "json",
-    },
-  );
-  const data = repoInfoRes.body as any;
-  return data;
-};
-
-export async function getProjectOptions(
-  owner: string,
-  repo: string,
-): Promise<
-  {
-    value: CommunityProjectConfig;
-    title: string;
-  }[]
-> {
-  // TODO: consider using octokit (https://github.com/octokit) if more changes are needed in the future
-  const getCommunityProjectConfig = async (
-    item: any,
-  ): Promise<CommunityProjectConfig | null> => {
-    // if item is a folder, return the path with default owner, repo, and main branch
-    if (item.type === "dir")
-      return {
-        owner,
-        repo,
-        branch: "main",
-        filePath: item.path,
-      };
-
-    // check if it's a submodule (has size = 0 and different owner & repo)
-    if (item.type === "file") {
-      if (item.size !== 0) return null; // submodules have size = 0
-
-      // get owner and repo from git_url
-      const { git_url } = item;
-      const startIndex = git_url.indexOf("repos/") + 6;
-      const endIndex = git_url.indexOf("/git");
-      const ownerRepoStr = git_url.substring(startIndex, endIndex);
-      const [owner, repo] = ownerRepoStr.split("/");
-
-      // quick fetch repo info to get the default branch
-      const { default_branch } = await getRepoInfo(owner, repo);
-
-      // return the path with default owner, repo, and main branch (path is empty for submodules)
-      return {
-        owner,
-        repo,
-        branch: default_branch,
-      };
-    }
-
-    return null;
-  };
-
-  const url = `https://api.github.com/repos/${owner}/${repo}/contents`;
-  const response = await got(url, {
-    responseType: "json",
-  });
-  const data = response.body as any[];
-
-  const projectConfigs: CommunityProjectConfig[] = [];
-  for (const item of data) {
-    const communityProjectConfig = await getCommunityProjectConfig(item);
-    if (communityProjectConfig) projectConfigs.push(communityProjectConfig);
-  }
-  return projectConfigs.map((config) => {
-    return {
-      value: config,
-      title: config.filePath || config.repo, // for submodules, use repo name as title
-    };
-  });
-}
-
-export async function getRepoRawContent(repoFilePath: string) {
-  const url = `https://raw.githubusercontent.com/${repoFilePath}`;
-  const response = await got(url, {
-    responseType: "text",
-  });
-  return response.body;
-}

helpers/tools.ts

@@ -1,340 +0,0 @@
-import fs from "fs/promises";
-import path from "path";
-import { red } from "picocolors";
-import yaml from "yaml";
-import { EnvVar } from "./env-variables";
-import { makeDir } from "./make-dir";
-import { TemplateFramework } from "./types";
-
-export const TOOL_SYSTEM_PROMPT_ENV_VAR = "TOOL_SYSTEM_PROMPT";
-
-export enum ToolType {
-  LLAMAHUB = "llamahub",
-  LOCAL = "local",
-}
-
-export type Tool = {
-  display: string;
-  name: string;
-  config?: Record<string, any>;
-  dependencies?: ToolDependencies[];
-  supportedFrameworks?: Array<TemplateFramework>;
-  type: ToolType;
-  envVars?: EnvVar[];
-};
-
-export type ToolDependencies = {
-  name: string;
-  version?: string;
-};
-
-export const supportedTools: Tool[] = [
-  {
-    display: "Google Search",
-    name: "google.GoogleSearchToolSpec",
-    config: {
-      engine:
-        "Your search engine id, see https://developers.google.com/custom-search/v1/overview#prerequisites",
-      key: "Your search api key",
-      num: 2,
-    },
-    dependencies: [
-      {
-        name: "llama-index-tools-google",
-        version: "^0.3.0",
-      },
-    ],
-    supportedFrameworks: ["fastapi"],
-    type: ToolType.LLAMAHUB,
-    envVars: [
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for google search tool.",
-        value: `You are a Google search agent. You help users to get information from Google search.`,
-      },
-    ],
-  },
-  {
-    // For python app, we will use a local DuckDuckGo search tool (instead of DuckDuckGo search tool in LlamaHub)
-    // to get the same results as the TS app.
-    display: "DuckDuckGo Search",
-    name: "duckduckgo",
-    dependencies: [
-      {
-        name: "duckduckgo-search",
-        version: "^6.3.5",
-      },
-    ],
-    supportedFrameworks: ["fastapi"], // TODO: Re-enable this tool once the duck-duck-scrape TypeScript library works again
-    type: ToolType.LOCAL,
-    envVars: [
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for DuckDuckGo search tool.",
-        value: `You have access to the duckduckgo search tool. Use it to get information from the web to answer user questions.
-For better results, you can specify the region parameter to get results from a specific region but it's optional.`,
-      },
-    ],
-  },
-  {
-    display: "Wikipedia",
-    name: "wikipedia.WikipediaToolSpec",
-    dependencies: [
-      {
-        name: "llama-index-tools-wikipedia",
-        version: "^0.3.0",
-      },
-    ],
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LLAMAHUB,
-  },
-  {
-    display: "Weather",
-    name: "weather",
-    dependencies: [],
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LOCAL,
-  },
-  {
-    display: "Document generator",
-    name: "document_generator",
-    supportedFrameworks: ["fastapi", "nextjs", "express"],
-    dependencies: [
-      {
-        name: "xhtml2pdf",
-        version: "^0.2.14",
-      },
-      {
-        name: "markdown",
-        version: "^3.7",
-      },
-    ],
-    type: ToolType.LOCAL,
-    envVars: [
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for document generator tool.",
-        value: `If user request for a report or a post, use document generator tool to create a file and reply with the link to the file.`,
-      },
-    ],
-  },
-  {
-    display: "Code Interpreter",
-    name: "interpreter",
-    dependencies: [
-      {
-        name: "e2b_code_interpreter",
-        version: "1.1.1",
-      },
-    ],
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LOCAL,
-    envVars: [
-      {
-        name: "E2B_API_KEY",
-        description:
-          "E2B_API_KEY key is required to run code interpreter tool. Get it here: https://e2b.dev/docs/getting-started/api-key",
-      },
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for code interpreter tool.",
-        value: `-You are a Python interpreter that can run any python code in a secure environment.
-- The python code runs in a Jupyter notebook. Every time you call the 'interpreter' tool, the python code is executed in a separate cell. 
-- You are given tasks to complete and you run python code to solve them.
-- It's okay to make multiple calls to interpreter tool. If you get an error or the result is not what you expected, you can call the tool again. Don't give up too soon!
-- Plot visualizations using matplotlib or any other visualization library directly in the notebook.
-- You can install any pip package (if it exists) by running a cell with pip install.`,
-      },
-    ],
-  },
-  {
-    display: "Artifact Code Generator",
-    name: "artifact",
-    // Using pre-release version of e2b_code_interpreter
-    // TODO: Update to stable version when 0.0.11 is released
-    dependencies: [
-      {
-        name: "e2b_code_interpreter",
-        version: "1.1.1",
-      },
-    ],
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LOCAL,
-    envVars: [
-      {
-        name: "E2B_API_KEY",
-        description:
-          "E2B_API_KEY key is required to run artifact code generator tool. Get it here: https://e2b.dev/docs/getting-started/api-key",
-      },
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for artifact code generator tool.",
-        value:
-          "You are a code assistant that can generate and execute code using its tools. Don't generate code yourself, use the provided tools instead. Do not show the code or sandbox url in chat, just describe the steps to build the application based on the code that is generated by your tools. Do not describe how to run the code, just the steps to build the application.",
-      },
-    ],
-  },
-  {
-    display: "OpenAPI action",
-    name: "openapi_action.OpenAPIActionToolSpec",
-    dependencies: [
-      {
-        name: "llama-index-tools-openapi",
-        version: "0.2.0",
-      },
-      {
-        name: "jsonschema",
-        version: "^4.22.0",
-      },
-      {
-        name: "llama-index-tools-requests",
-        version: "0.2.0",
-      },
-    ],
-    config: {
-      openapi_uri: "The URL or file path of the OpenAPI schema",
-    },
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LOCAL,
-  },
-  {
-    display: "Image Generator",
-    name: "img_gen",
-    supportedFrameworks: ["fastapi", "express", "nextjs"],
-    type: ToolType.LOCAL,
-    envVars: [
-      {
-        name: "STABILITY_API_KEY",
-        description:
-          "STABILITY_API_KEY key is required to run image generator. Get it here: https://platform.stability.ai/account/keys",
-      },
-    ],
-  },
-  {
-    display: "Azure Code Interpreter",
-    name: "azure_code_interpreter.AzureCodeInterpreterToolSpec",
-    supportedFrameworks: ["fastapi", "nextjs", "express"],
-    type: ToolType.LLAMAHUB,
-    dependencies: [
-      {
-        name: "llama-index-tools-azure-code-interpreter",
-        version: "0.2.0",
-      },
-    ],
-    envVars: [
-      {
-        name: "AZURE_POOL_MANAGEMENT_ENDPOINT",
-        description:
-          "Please follow this guideline to create and get the pool management endpoint: https://learn.microsoft.com/azure/container-apps/sessions?tabs=azure-cli",
-      },
-      {
-        name: TOOL_SYSTEM_PROMPT_ENV_VAR,
-        description: "System prompt for Azure code interpreter tool.",
-        value: `-You are a Python interpreter that can run any python code in a secure environment.
-- The python code runs in a Jupyter notebook. Every time you call the 'interpreter' tool, the python code is executed in a separate cell. 
-- You are given tasks to complete and you run python code to solve them.
-- It's okay to make multiple calls to interpreter tool. If you get an error or the result is not what you expected, you can call the tool again. Don't give up too soon!
-- Plot visualizations using matplotlib or any other visualization library directly in the notebook.
-- You can install any pip package (if it exists) by running a cell with pip install.`,
-      },
-    ],
-  },
-  {
-    display: "Form Filling",
-    name: "form_filling",
-    supportedFrameworks: ["fastapi"],
-    type: ToolType.LOCAL,
-    dependencies: [
-      {
-        name: "pandas",
-        version: "^2.2.3",
-      },
-      {
-        name: "tabulate",
-        version: "^0.9.0",
-      },
-    ],
-  },
-];
-
-export const getTool = (toolName: string): Tool | undefined => {
-  return supportedTools.find((tool) => tool.name === toolName);
-};
-
-export const getTools = (toolsName: string[]): Tool[] => {
-  const tools: Tool[] = [];
-  for (const toolName of toolsName) {
-    const tool = getTool(toolName);
-    if (!tool) {
-      console.log(
-        red(
-          `Error: Tool '${toolName}' is not supported. Supported tools are: ${supportedTools
-            .map((t) => t.name)
-            .join(", ")}`,
-        ),
-      );
-      process.exit(1);
-    }
-    tools.push(tool);
-  }
-  return tools;
-};
-
-export const toolRequiresConfig = (tool: Tool): boolean => {
-  const hasConfig = Object.keys(tool.config || {}).length > 0;
-  const hasEmptyEnvVar = tool.envVars?.some((envVar) => !envVar.value) ?? false;
-  return hasConfig || hasEmptyEnvVar;
-};
-
-export const toolsRequireConfig = (tools?: Tool[]): boolean => {
-  if (tools) {
-    return tools?.some(toolRequiresConfig);
-  }
-  return false;
-};
-
-export enum ConfigFileType {
-  YAML = "yaml",
-  JSON = "json",
-}
-
-export const writeToolsConfig = async (
-  root: string,
-  tools: Tool[] = [],
-  type: ConfigFileType = ConfigFileType.YAML,
-) => {
-  const configContent: {
-    [key in ToolType]: Record<string, any>;
-  } = {
-    local: {},
-    llamahub: {},
-  };
-  tools.forEach((tool) => {
-    if (tool.type === ToolType.LLAMAHUB) {
-      configContent.llamahub[tool.name] = tool.config ?? {};
-    }
-    if (tool.type === ToolType.LOCAL) {
-      configContent.local[tool.name] = tool.config ?? {};
-    }
-  });
-  const configPath = path.join(root, "config");
-  await makeDir(configPath);
-  if (type === ConfigFileType.YAML) {
-    await fs.writeFile(
-      path.join(configPath, "tools.yaml"),
-      yaml.stringify(configContent),
-    );
-  } else {
-    // For Typescript, we treat llamahub tools as local tools
-    const tsConfigContent = {
-      local: {
-        ...configContent.local,
-        ...configContent.llamahub,
-      },
-    };
-    await fs.writeFile(
-      path.join(configPath, "tools.json"),
-      JSON.stringify(tsConfigContent, null, 2),
-    );
-  }
-};

helpers/typescript.ts

@@ -1,570 +0,0 @@
-import fs from "fs/promises";
-import os from "os";
-import path from "path";
-import { bold, cyan, red, yellow } from "picocolors";
-import { assetRelocator, copy } from "../helpers/copy";
-import { callPackageManager } from "../helpers/install";
-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,
-    ),
-  });
-
-  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 ({
-  root,
-  template,
-  backend,
-  framework,
-  ui,
-  vectorDb,
-  observability,
-  tools,
-  dataSources,
-  useLlamaParse,
-  useCase,
-  modelConfig,
-  relativeEngineDestPath,
-}: InstallTemplateArgs & {
-  backend: boolean;
-  relativeEngineDestPath: string;
-}) => {
-  /**
-   * If next.js is used, update its configuration if necessary
-   */
-  if (framework === "nextjs") {
-    const nextConfigJsonFile = path.join(root, "next.config.json");
-    const nextConfigJson: any = JSON.parse(
-      await fs.readFile(nextConfigJsonFile, "utf8"),
-    );
-    if (!backend) {
-      // update next.config.json for static site generation
-      nextConfigJson.output = "export";
-      nextConfigJson.images = { unoptimized: true };
-      console.log("\nUsing static site generation\n");
-    } else {
-      if (vectorDb === "milvus") {
-        nextConfigJson.serverExternalPackages =
-          nextConfigJson.serverExternalPackages ?? [];
-        nextConfigJson.serverExternalPackages.push("@zilliz/milvus2-sdk-node");
-      }
-    }
-    await fs.writeFile(
-      nextConfigJsonFile,
-      JSON.stringify(nextConfigJson, null, 2) + os.EOL,
-    );
-
-    const webpackConfigOtelFile = path.join(root, "webpack.config.o11y.mjs");
-    if (observability === "traceloop") {
-      const webpackConfigDefaultFile = path.join(root, "webpack.config.mjs");
-      await fs.rm(webpackConfigDefaultFile);
-      await fs.rename(webpackConfigOtelFile, webpackConfigDefaultFile);
-    } else {
-      await fs.rm(webpackConfigOtelFile);
-    }
-  }
-
-  // copy observability component
-  if (observability && observability !== "none") {
-    const chosenObservabilityPath = path.join(
-      templatesDir,
-      "components",
-      "observability",
-      "typescript",
-      observability,
-    );
-    const relativeObservabilityPath = framework === "nextjs" ? "app" : "src";
-
-    await copy(
-      "**",
-      path.join(root, relativeObservabilityPath, "observability"),
-      { cwd: chosenObservabilityPath },
-    );
-  }
-
-  const compPath = path.join(templatesDir, "components");
-  const enginePath = path.join(root, relativeEngineDestPath, "engine");
-
-  // copy llamaindex code for TS templates
-  await copy("**", path.join(root, relativeEngineDestPath, "llamaindex"), {
-    parents: true,
-    cwd: path.join(compPath, "llamaindex", "typescript"),
-  });
-
-  // copy vector db component
-  if (vectorDb === "llamacloud") {
-    console.log(
-      `\nUsing managed index from LlamaCloud. Ensure the ${yellow("LLAMA_CLOUD_* environment variables are set correctly.")}`,
-    );
-  } else {
-    console.log("\nUsing vector DB:", vectorDb ?? "none");
-  }
-  await copy("**", enginePath, {
-    parents: true,
-    cwd: path.join(compPath, "vectordbs", "typescript", vectorDb ?? "none"),
-  });
-
-  if (template === "multiagent") {
-    const multiagentPath = path.join(compPath, "multiagent", "typescript");
-
-    // copy workflow code for multiagent template
-    await copy("**", path.join(root, relativeEngineDestPath, "workflow"), {
-      parents: true,
-      cwd: path.join(multiagentPath, "workflow"),
-    });
-
-    // Copy use case code for multiagent template
-    if (useCase) {
-      console.log("\nCopying use case:", useCase, "\n");
-      const useCasePath = path.join(compPath, "agents", "typescript", useCase);
-      const useCaseCodePath = path.join(useCasePath, "workflow");
-
-      // Copy use case codes
-      await copy("**", path.join(root, relativeEngineDestPath, "workflow"), {
-        parents: true,
-        cwd: useCaseCodePath,
-        rename: assetRelocator,
-      });
-
-      // Copy use case files to project root
-      await copy("*.*", path.join(root), {
-        parents: true,
-        cwd: useCasePath,
-        rename: assetRelocator,
-      });
-    } else {
-      console.log(
-        red(
-          `There is no use case selected for ${template} template. Please pick a use case to use via --use-case flag.`,
-        ),
-      );
-      process.exit(1);
-    }
-
-    if (framework === "nextjs") {
-      // patch route.ts file
-      await copy("**", path.join(root, relativeEngineDestPath), {
-        parents: true,
-        cwd: path.join(multiagentPath, "nextjs"),
-      });
-    } else if (framework === "express") {
-      // patch chat.controller.ts file
-      await copy("**", path.join(root, relativeEngineDestPath), {
-        parents: true,
-        cwd: path.join(multiagentPath, "express"),
-      });
-    }
-  }
-
-  // copy loader component (TS only supports llama_parse and file for now)
-  const loaderFolder = useLlamaParse ? "llama_parse" : "file";
-  await copy("**", enginePath, {
-    parents: true,
-    cwd: path.join(compPath, "loaders", "typescript", loaderFolder),
-  });
-
-  // copy provider settings
-  await copy("**", enginePath, {
-    parents: true,
-    cwd: path.join(compPath, "providers", "typescript", modelConfig.provider),
-  });
-
-  // Select and copy engine code based on data sources and tools
-  let engine;
-  tools = tools ?? [];
-  // multiagent template always uses agent engine
-  if (template === "multiagent") {
-    engine = "agent";
-  } else if (dataSources.length > 0 && tools.length === 0) {
-    console.log("\nNo tools selected - use optimized context chat engine\n");
-    engine = "chat";
-  } else {
-    engine = "agent";
-  }
-  await copy("**", enginePath, {
-    parents: true,
-    cwd: path.join(compPath, "engines", "typescript", engine),
-  });
-
-  // copy settings to engine folder
-  await copy("**", enginePath, {
-    cwd: path.join(compPath, "settings", "typescript"),
-  });
-
-  /**
-   * Copy the selected UI files to the target directory and reference it.
-   */
-  if (framework === "nextjs" && ui !== "shadcn") {
-    console.log("\nUsing UI:", ui, "\n");
-    const uiPath = path.join(compPath, "ui", ui);
-    const destUiPath = path.join(root, "app", "components", "ui");
-    // remove the default ui folder
-    await fs.rm(destUiPath, { recursive: true });
-    // copy the selected ui folder
-    await copy("**", destUiPath, {
-      parents: true,
-      cwd: uiPath,
-      rename: assetRelocator,
-    });
-  }
-
-  /** Modify frontend code to use custom API path */
-  if (framework === "nextjs" && !backend) {
-    console.log(
-      "\nUsing external API for frontend, removing API code and configuration\n",
-    );
-    // remove the default api folder and config folder
-    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,
-    appName,
-    dataSources,
-    relativeEngineDestPath,
-    framework,
-    ui,
-    observability,
-    vectorDb,
-    backend,
-    modelConfig,
-    template,
-  });
-
-  if (
-    backend &&
-    (postInstallAction === "runApp" || postInstallAction === "dependencies")
-  ) {
-    await installTSDependencies(packageJson, packageManager, isOnline);
-  }
-};
-
-const providerDependencies: {
-  [key in ModelProvider]?: Record<string, string>;
-} = {
-  openai: {
-    "@llamaindex/openai": "^0.2.0",
-  },
-  gemini: {
-    "@llamaindex/google": "^0.2.0",
-  },
-  ollama: {
-    "@llamaindex/ollama": "^0.1.0",
-  },
-  mistral: {
-    "@llamaindex/mistral": "^0.2.0",
-  },
-  "azure-openai": {
-    "@llamaindex/openai": "^0.2.0",
-  },
-  groq: {
-    "@llamaindex/groq": "^0.0.61",
-    "@llamaindex/huggingface": "^0.1.0", // 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
-  },
-};
-
-const vectorDbDependencies: Record<TemplateVectorDB, Record<string, string>> = {
-  astra: {
-    "@llamaindex/astra": "^0.0.5",
-  },
-  chroma: {
-    "@llamaindex/chroma": "^0.0.5",
-  },
-  llamacloud: {},
-  milvus: {
-    "@zilliz/milvus2-sdk-node": "^2.4.6",
-    "@llamaindex/milvus": "^0.1.0",
-  },
-  mongo: {
-    mongodb: "6.7.0",
-    "@llamaindex/mongodb": "^0.0.5",
-  },
-  none: {},
-  pg: {
-    pg: "^8.12.0",
-    pgvector: "^0.2.0",
-    "@llamaindex/postgres": "^0.0.33",
-  },
-  pinecone: {
-    "@llamaindex/pinecone": "^0.0.5",
-  },
-  qdrant: {
-    "@qdrant/js-client-rest": "^1.11.0",
-    "@llamaindex/qdrant": "^0.1.0",
-  },
-  weaviate: {
-    "@llamaindex/weaviate": "^0.0.5",
-  },
-};
-
-async function updatePackageJson({
-  root,
-  appName,
-  dataSources,
-  relativeEngineDestPath,
-  framework,
-  ui,
-  observability,
-  vectorDb,
-  backend,
-  modelConfig,
-  template,
-}: Pick<
-  InstallTemplateArgs,
-  | "root"
-  | "appName"
-  | "dataSources"
-  | "framework"
-  | "ui"
-  | "observability"
-  | "vectorDb"
-  | "modelConfig"
-  | "template"
-> & {
-  relativeEngineDestPath: string;
-  backend: boolean;
-}): Promise<any> {
-  const packageJsonFile = path.join(root, "package.json");
-  const packageJson: any = JSON.parse(
-    await fs.readFile(packageJsonFile, "utf8"),
-  );
-  packageJson.name = appName;
-  packageJson.version = "0.1.0";
-
-  if (relativeEngineDestPath && template !== "llamaindexserver") {
-    // TODO: move script to {root}/scripts for all frameworks
-    // add generate script if using context engine
-    packageJson.scripts = {
-      ...packageJson.scripts,
-      generate: `tsx ${path.join(
-        relativeEngineDestPath,
-        "engine",
-        "generate.ts",
-      )}`,
-    };
-  }
-
-  if (framework === "nextjs" && ui === "html") {
-    // remove shadcn dependencies if html ui is selected
-    packageJson.dependencies = {
-      ...packageJson.dependencies,
-      "tailwind-merge": undefined,
-      "@radix-ui/react-slot": undefined,
-      "class-variance-authority": undefined,
-      clsx: undefined,
-      "lucide-react": undefined,
-      remark: undefined,
-      "remark-code-import": undefined,
-      "remark-gfm": undefined,
-      "remark-math": undefined,
-      "react-markdown": undefined,
-      "highlight.js": undefined,
-    };
-  }
-
-  if (backend) {
-    packageJson.dependencies = {
-      ...packageJson.dependencies,
-      "@llamaindex/readers": "^2.0.0",
-    };
-
-    if (vectorDb && vectorDb in vectorDbDependencies) {
-      packageJson.dependencies = {
-        ...packageJson.dependencies,
-        ...vectorDbDependencies[vectorDb],
-      };
-    }
-
-    if (modelConfig.provider && modelConfig.provider in providerDependencies) {
-      packageJson.dependencies = {
-        ...packageJson.dependencies,
-        ...providerDependencies[modelConfig.provider],
-      };
-    }
-  }
-
-  if (observability === "traceloop") {
-    packageJson.dependencies = {
-      ...packageJson.dependencies,
-      "@traceloop/node-server-sdk": "^0.5.19",
-    };
-
-    packageJson.devDependencies = {
-      ...packageJson.devDependencies,
-      "node-loader": "^2.0.0",
-    };
-  }
-
-  await fs.writeFile(
-    packageJsonFile,
-    JSON.stringify(packageJson, null, 2) + os.EOL,
-  );
-
-  return packageJson;
-}
-
-async function installTSDependencies(
-  packageJson: any,
-  packageManager: PackageManager,
-  isOnline: boolean,
-): Promise<void> {
-  console.log("\nInstalling dependencies:");
-  for (const dependency in packageJson.dependencies)
-    console.log(`- ${cyan(dependency)}`);
-
-  console.log("\nInstalling devDependencies:");
-  for (const dependency in packageJson.devDependencies)
-    console.log(`- ${cyan(dependency)}`);
-
-  console.log();
-
-  await callPackageManager(packageManager, isOnline).catch((error) => {
-    console.error("Failed to install TS dependencies. Exiting...");
-    process.exit(1);
-  });
-}

llama-index-server/README.md

@@ -1,130 +0,0 @@
-# LlamaIndex Server
-
-LlamaIndexServer is a FastAPI-based application that allows you to quickly launch your [LlamaIndex Workflows](https://docs.llamaindex.ai/en/stable/module_guides/workflow/#workflows) and [Agent Workflows](https://docs.llamaindex.ai/en/stable/understanding/agent/multi_agent/) as an API server with an optional chat UI. It provides a complete environment for running LlamaIndex workflows with both API endpoints and a user interface for interaction.
-
-## Features
-
-- Serving a workflow as a chatbot
-- Built on FastAPI for high performance and easy API development
-- Optional built-in chat UI
-- Prebuilt development code
-
-## Installation
-
-```bash
-pip install llama-index-server
-```
-
-## Quick Start
-
-```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 for the workflow
-app = LlamaIndexServer(
-    workflow_factory=create_workflow,  # Supports Workflow or AgentWorkflow
-    env="dev",  # Enable development mode
-    include_ui=True,  # Include chat UI
-    starter_questions=["What can you do?", "How do I use this?"],
-    verbose=True
-)
-```
-
-## 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`
-- Access the chat UI at `http://localhost:8000/` (Make sure you set the `env="dev"` or `include_ui=True` in the server configuration)
-
-## Configuration Options
-
-The LlamaIndexServer accepts the following configuration parameters:
-
-- `workflow_factory`: A callable that creates a workflow instance for each request
-- `logger`: Optional logger instance (defaults to uvicorn logger)
-- `use_default_routers`: Whether to include default routers (chat, static file serving)
-- `env`: Environment setting ('dev' enables CORS and UI by default)
-- `include_ui`: Whether to include the chat UI
-- `starter_questions`: List of starter questions for the chat UI
-- `verbose`: Enable verbose logging
-- `api_prefix`: API route prefix (default: "/api")
-- `server_url`: The deployment URL of the server (default is None)
-- `ui_path`: Path for downloaded UI static files (default: ".ui")
-
-## Default Routers and Features
-
-### Chat Router
-
-The server includes a default chat router at `/api/chat` for handling chat interactions.
-
-### Static File Serving
-
-- The server automatically mounts the `data` and `output` folders at `{server_url}{api_prefix}/files/data` (default: `/api/files/data`) and `{server_url}{api_prefix}/files/output` (default: `/api/files/output`) respectively.
-- Your workflows can use both folders to store and access files. As a convention, the `data` folder is used for documents that are ingested and the `output` folder is used for documents that are generated by the workflow.
-- The example workflows from `create-llama` (see below) are following this pattern.
-
-### Chat UI
-
-When enabled, the server provides a chat interface at the root path (`/`) with:
-
-- Configurable starter questions
-- Real-time chat interface
-- API endpoint integration
-
-## Development Mode
-
-In development mode (`env="dev"`), the server:
-
-- Enables CORS for all origins
-- Automatically includes the chat UI
-- Provides more verbose logging
-
-## API Endpoints
-
-The server provides the following default endpoints:
-
-- `/api/chat`: Chat interaction endpoint
-- `/api/files/data/*`: Access to data directory files
-- `/api/files/output/*`: Access to output directory files
-
-## Best Practices
-
-1. Always provide a workflow factory that creates fresh workflow instances
-2. Use environment variables for sensitive configuration
-3. Enable verbose logging during development
-4. Configure CORS appropriately for your deployment environment
-5. Use starter questions to guide users in the chat UI
-
-## Getting Started with a New Project
-
-Want to start a new project with LlamaIndexServer? Check out our [create-llama](https://github.com/run-llama/create-llama) tool to quickly generate a new project with LlamaIndexServer.

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

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

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

@@ -1,32 +0,0 @@
-from typing import Any
-
-from llama_index.core.agent.workflow.workflow_events import ToolCallResult
-from llama_index.server.api.callbacks.base import EventCallback
-from llama_index.server.api.models import SourceNodesEvent
-
-
-class SourceNodesFromToolCall(EventCallback):
-    """
-    Extract source nodes from the query tool output.
-
-    Args:
-        query_tool_name: The name of the tool that queries the index.
-                         default is "query_index"
-    """
-
-    def __init__(self, query_tool_name: str = "query_index"):
-        self.query_tool_name = query_tool_name
-
-    def transform_tool_call_result(self, event: ToolCallResult) -> SourceNodesEvent:
-        source_nodes = event.tool_output.raw_output.source_nodes
-        return SourceNodesEvent(nodes=source_nodes)
-
-    async def run(self, event: Any) -> Any:
-        if isinstance(event, ToolCallResult):
-            if event.tool_name == self.query_tool_name:
-                return event, self.transform_tool_call_result(event)
-        return event
-
-    @classmethod
-    def from_default(cls, *args: Any, **kwargs: Any) -> "SourceNodesFromToolCall":
-        return cls()

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

@@ -1,136 +0,0 @@
-import logging
-import os
-from enum import Enum
-from typing import Any, Dict, List, Optional
-
-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
-from pydantic import BaseModel, Field, field_validator
-
-logger = logging.getLogger("uvicorn")
-
-
-class ChatConfig(BaseModel):
-    next_question_suggestions: bool = Field(
-        default=True,
-        description="Whether to suggest next questions",
-    )
-
-
-class ChatAPIMessage(BaseModel):
-    role: MessageRole
-    content: str
-
-    def to_llamaindex_message(self) -> ChatMessage:
-        return ChatMessage(role=self.role, content=self.content)
-
-
-class ChatRequest(BaseModel):
-    messages: List[ChatAPIMessage]
-    data: Optional[Any] = None
-    config: Optional[ChatConfig] = ChatConfig()
-
-    @field_validator("messages")
-    def validate_messages(cls, v: List[ChatAPIMessage]) -> List[ChatAPIMessage]:
-        if v[-1].role != MessageRole.USER:
-            raise ValueError("Last message must be from user")
-        return v
-
-
-class AgentRunEventType(Enum):
-    TEXT = "text"
-    PROGRESS = "progress"
-
-
-class AgentRunEvent(Event):
-    name: str
-    msg: str
-    event_type: AgentRunEventType = AgentRunEventType.TEXT
-    data: Optional[dict] = None
-
-    def to_response(self) -> dict:
-        return {
-            "type": "agent",
-            "data": {
-                "agent": self.name,
-                "type": self.event_type.value,
-                "text": self.msg,
-                "data": self.data,
-            },
-        }
-
-
-class SourceNodesEvent(Event):
-    nodes: List[NodeWithScore]
-
-    def to_response(self) -> dict:
-        return {
-            "type": "sources",
-            "data": {
-                "nodes": [
-                    SourceNodes.from_source_node(node).model_dump()
-                    for node in self.nodes
-                ]
-            },
-        }
-
-
-class SourceNodes(BaseModel):
-    id: str
-    metadata: Dict[str, Any]
-    score: Optional[float]
-    text: str
-    url: Optional[str]
-
-    @classmethod
-    def from_source_node(cls, source_node: NodeWithScore) -> "SourceNodes":
-        metadata = source_node.node.metadata
-        url = cls.get_url_from_metadata(metadata)
-
-        return cls(
-            id=source_node.node.node_id,
-            metadata=metadata,
-            score=source_node.score,
-            text=source_node.node.text,  # type: ignore
-            url=url,
-        )
-
-    @classmethod
-    def get_url_from_metadata(
-        cls,
-        metadata: Dict[str, Any],
-        data_dir: Optional[str] = None,
-    ) -> Optional[str]:
-        url_prefix = server_settings.file_server_url_prefix
-        if data_dir is None:
-            data_dir = "data"
-        file_name = metadata.get("file_name")
-
-        if file_name and url_prefix:
-            # file_name exists and file server is configured
-            pipeline_id = metadata.get("pipeline_id")
-            if pipeline_id:
-                # file is from LlamaCloud
-                file_name = f"{pipeline_id}${file_name}"
-                return f"{url_prefix}/output/llamacloud/{file_name}"
-            is_private = metadata.get("private", "false") == "true"
-            if is_private:
-                # file is a private upload
-                return f"{url_prefix}/output/uploaded/{file_name}"
-            # file is from calling the 'generate' script
-            # Get the relative path of file_path to data_dir
-            file_path = metadata.get("file_path")
-            data_dir = os.path.abspath(data_dir)
-            if file_path and data_dir:
-                relative_path = os.path.relpath(file_path, data_dir)
-                return f"{url_prefix}/data/{relative_path}"
-        # fallback to URL in metadata (e.g. for websites)
-        return metadata.get("URL")
-
-    @classmethod
-    def from_source_nodes(
-        cls, source_nodes: List[NodeWithScore]
-    ) -> List["SourceNodes"]:
-        return [cls.from_source_node(node) for node in source_nodes]

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

@@ -1,55 +0,0 @@
-import logging
-import shutil
-from pathlib import Path
-from typing import Optional
-
-import requests
-
-CHAT_UI_VERSION = "0.0.6"
-
-
-def download_chat_ui(
-    logger: Optional[logging.Logger] = None, target_path: str = ".ui"
-) -> None:
-    if logger is None:
-        logger = logging.getLogger("uvicorn")
-    path = Path(target_path)
-    temp_dir = _download_package(_get_download_link(CHAT_UI_VERSION))
-    _copy_ui_files(temp_dir, path)
-    logger.info("Chat UI downloaded and copied to static folder")
-
-
-def _get_download_link(version: str) -> str:
-    """Get the download link for the chat UI from the npm registry."""
-    return f"https://registry.npmjs.org/@llamaindex/server/-/server-{version}.tgz"
-
-
-def _download_package(url: str) -> Path:
-    """Download tar.gz file and extract all files into a temporary directory."""
-    import io
-    import tarfile
-    import tempfile
-
-    response = requests.get(url, headers={"User-Agent": "Mozilla/5.0"})
-    content = response.content
-
-    temp_dir = Path(tempfile.mkdtemp())
-
-    with tarfile.open(fileobj=io.BytesIO(content), mode="r:gz") as tar:
-        tar.extractall(path=temp_dir)
-
-    return temp_dir
-
-
-def _copy_ui_files(temp_dir: Path, target_path: Path) -> None:
-    """Copy files from the .next directory to the static directory."""
-    target_path.mkdir(parents=True, exist_ok=True)
-    next_dir = temp_dir / "package/dist/static"
-
-    if next_dir.exists():
-        for item in next_dir.iterdir():
-            dest = target_path / item.name
-            if item.is_dir():
-                shutil.copytree(item, dest, dirs_exist_ok=True)
-            else:
-                shutil.copy2(item, dest)

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

@@ -1,184 +0,0 @@
-import json
-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
-from llama_index.server.settings import server_settings
-
-
-class LlamaIndexServer(FastAPI):
-    workflow_factory: Callable[..., Workflow]
-    include_ui: Optional[bool]
-    starter_questions: Optional[list[str]]
-    verbose: bool = False
-    ui_path: str = ".ui"
-
-    def __init__(
-        self,
-        workflow_factory: Callable[..., Workflow],
-        logger: Optional[logging.Logger] = None,
-        use_default_routers: Optional[bool] = True,
-        env: Optional[str] = None,
-        include_ui: Optional[bool] = None,
-        starter_questions: Optional[list[str]] = None,
-        server_url: Optional[str] = None,
-        api_prefix: Optional[str] = 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.
-            starter_questions: A list of starter questions to display in the chat UI.
-            server_url: The URL of the server.
-            api_prefix: The prefix for the API endpoints.
-            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
-        self.starter_questions = starter_questions
-        self.use_default_routers = use_default_routers or True
-
-        # Update the settings
-        if server_url:
-            server_settings.set_url(server_url)
-        if api_prefix:
-            server_settings.set_api_prefix(api_prefix)
-
-        if self.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()
-
-    @property
-    def _ui_config(self) -> dict:
-        config = {
-            "CHAT_API": f"{server_settings.api_url}/chat",
-            "STARTER_QUESTIONS": self.starter_questions,
-        }
-        is_llamacloud_configured = os.getenv("LLAMA_CLOUD_API_KEY") is not None
-        if is_llamacloud_configured:
-            config["LLAMA_CLOUD_API"] = (
-                f"{server_settings.api_url}/chat/config/llamacloud"
-            )
-        return config
-
-    # 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=server_settings.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)
-            self._override_ui_config()
-
-    def _override_ui_config(self) -> None:
-        """
-        Override the UI config by writing a complete configuration file.
-        """
-        try:
-            config_path = os.path.join(self.ui_path, "config.js")
-            if not os.path.exists(config_path):
-                self.logger.error("Config file not found")
-                return
-            config_content = (
-                f"window.LLAMAINDEX = {json.dumps(self._ui_config, indent=2)};"
-            )
-            with open(config_path, "w") as f:
-                f.write(config_content)
-        except Exception as e:
-            self.logger.error(f"Error overriding UI config: {e}")
-
-    def mount_data_dir(self, data_dir: str = "data") -> None:
-        """
-        Mount the data directory.
-        """
-        self._mount_static_files(
-            directory=data_dir,
-            path=f"{server_settings.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"{server_settings.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

@@ -1,117 +0,0 @@
-import logging
-import os
-import re
-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__)
-
-PRIVATE_STORE_PATH = str(Path("output", "uploaded"))
-TOOL_STORE_PATH = str(Path("output", "tools"))
-LLAMA_CLOUD_STORE_PATH = str(Path("output", "llamacloud"))
-
-
-class DocumentFile(BaseModel):
-    id: str
-    name: str  # Stored file name
-    type: Optional[str] = None
-    size: Optional[int] = None
-    url: Optional[str] = None
-    path: Optional[str] = Field(
-        None,
-        description="The stored file path. Used internally in the server.",
-        exclude=True,
-    )
-    refs: Optional[List[str]] = Field(
-        None, description="The document ids in the index."
-    )
-
-
-class FileService:
-    """
-    To store the files uploaded by the user.
-    """
-
-    @classmethod
-    def save_file(
-        cls,
-        content: Union[bytes, str],
-        file_name: str,
-        save_dir: Optional[str] = None,
-    ) -> DocumentFile:
-        """
-        Save the content to a file in the local file server (accessible via URL).
-
-        Args:
-            content (bytes | str): The content to save, either bytes or string.
-            file_name (str): The original name of the file.
-            save_dir (Optional[str]): The relative path from the current working directory. Defaults to the `output/uploaded` directory.
-
-        Returns:
-            The metadata of the saved file.
-        """
-        if save_dir is None:
-            save_dir = os.path.join("output", "uploaded")
-
-        file_id = str(uuid.uuid4())
-        name, extension = os.path.splitext(file_name)
-        extension = extension.lstrip(".")
-        sanitized_name = _sanitize_file_name(name)
-        if extension == "":
-            raise ValueError("File is not supported!")
-        new_file_name = f"{sanitized_name}_{file_id}.{extension}"
-
-        file_path = os.path.join(save_dir, new_file_name)
-
-        if isinstance(content, str):
-            content = content.encode()
-
-        try:
-            os.makedirs(os.path.dirname(file_path), exist_ok=True)
-            with open(file_path, "wb") as file:
-                file.write(content)
-        except PermissionError as e:
-            logger.error(f"Permission denied when writing to file {file_path}: {e!s}")
-            raise
-        except OSError as e:
-            logger.error(f"IO error occurred when writing to file {file_path}: {e!s}")
-            raise
-        except Exception as e:
-            logger.error(f"Unexpected error when writing to file {file_path}: {e!s}")
-            raise
-
-        logger.info(f"Saved file to {file_path}")
-
-        file_size = os.path.getsize(file_path)
-        file_url = (
-            f"{server_settings.file_server_url_prefix}/{save_dir}/{new_file_name}"
-        )
-        return DocumentFile(
-            id=file_id,
-            name=new_file_name,
-            type=extension,
-            size=file_size,
-            path=file_path,
-            url=file_url,
-            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:
-    """
-    Sanitize the file name by replacing all non-alphanumeric characters with underscores.
-    """
-    return re.sub(r"[^a-zA-Z0-9.]", "_", file_name)

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

@@ -1,3 +0,0 @@
-from .query import get_query_engine_tool
-
-__all__ = ["get_query_engine_tool"]

llama-index-server/pyproject.toml

@@ -1,64 +0,0 @@
-[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 = "0.1.7"
-
-[tool.poetry.dependencies]
-python = ">=3.9,<4.0"
-fastapi = {extras = ["standard"], version = "^0.115.11"}
-cachetools = "^5.5.2"
-requests = "^2.32.3"
-pydantic-settings = "^2.8.1"
-llama-index-core = "0.12.25"
-llama-index-readers-file = "^0.4.6"
-llama-index-indices-managed-llama-cloud = "0.6.3"
-
-[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-cloud = "^0.1.17"

llama-index-server/tests/test_llamaindex_server.py

@@ -1,106 +0,0 @@
-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"]

package.json

@@ -1,83 +1,55 @@
 {
-  "name": "create-llama",
-  "version": "0.5.0",
-  "description": "Create LlamaIndex-powered apps with one command",
+  "name": "create-llama-monorepo",
+  "version": "1.0.0",
+  "private": true,
+  "description": "Monorepo for create-llama",
   "keywords": [
     "rag",
-    "llamaindex",
-    "next.js"
+    "llamaindex"
   ],
   "repository": {
     "type": "git",
-    "url": "https://github.com/run-llama/create-llama",
-    "directory": "packages/create-llama"
+    "url": "https://github.com/run-llama/create-llama"
   },
   "license": "MIT",
-  "bin": {
-    "create-llama": "./dist/index.js"
-  },
-  "files": [
-    "dist"
+  "workspaces": [
+    "packages/*",
+    "python/*"
   ],
   "scripts": {
-    "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",
+    "dev": "pnpm -r dev",
+    "build": "pnpm -r build",
+    "e2e": "pnpm -r e2e",
+    "lint": "eslint .",
     "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",
-    "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"
+    "new-snapshot": "pnpm -r build && changeset version --snapshot",
+    "new-version-python": "pnpm --filter @create-llama/llama-index-server new-version",
+    "new-version": "pnpm -r build && changeset version && pnpm new-version-python",
+    "release-python": "pnpm --filter @create-llama/llama-index-server release",
+    "release": "pnpm -r build && changeset publish && pnpm release-python",
+    "release-snapshot": "pnpm -r build && changeset publish --tag snapshot"
   },
   "devDependencies": {
     "@changesets/cli": "^2.27.1",
-    "@playwright/test": "^1.41.1",
-    "@vercel/ncc": "0.38.1",
-    "eslint": "^8.56.0",
-    "eslint-config-prettier": "^8.10.0",
+    "bunchee": "6.4.0",
     "husky": "^9.0.10",
-    "prettier": "^3.2.5",
-    "prettier-plugin-organize-imports": "^3.2.4",
-    "rimraf": "^5.0.5",
-    "typescript": "^5.3.3",
-    "wait-port": "^1.1.0"
+    "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"
   },
   "packageManager": "pnpm@9.0.5",
   "engines": {

packages/create-llama/.gitignore

@@ -0,0 +1,65 @@
+# 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/CHANGELOG.md

@@ -1,5 +1,165 @@
 # create-llama
 
+## 0.6.1
+
+### Patch Changes
+
+- 952b5b4: fix: peer deps and sourcemap issues made ts server start fail
+- e8004fd: Fix: broken devcontainer due to deleted repo
+
+## 0.6.0
+
+### Minor Changes
+
+- 8fa8c3b: Removed deprecated templates and simplified code
+
+### Patch Changes
+
+- 8fa8c3b: Feat: re-add --ask-models
+
+## 0.5.22
+
+### Patch Changes
+
+- e2486eb: feat: support human in the loop for TS
+
+## 0.5.21
+
+### Patch Changes
+
+- af9ad3c: feat: show document artifact after generating report
+- a543a27: feat: bump chat-ui with inline artifact
+
+## 0.5.20
+
+### Patch Changes
+
+- 3ff0a18: fix: default header padding
+
+## 0.5.19
+
+### Patch Changes
+
+- 5fe9e17: support eject to fully customize next folder
+- b8a1ff6: Support citation for agentic template (Python)
+
+## 0.5.18
+
+### Patch Changes
+
+- 8d59ef0: Add layout_dir config to the generated python code
+
+## 0.5.17
+
+### Patch Changes
+
+- eee3230: feat: support custom layout
+
+## 0.5.16
+
+### Patch Changes
+
+- 6f75d4a: fix: unsupported language in code gen workflow
+- d0618fa: Fix LlamaCloud generate script issue
+
+## 0.5.15
+
+### Patch Changes
+
+- 527075c: Enable dev mode that allows updating code directly in the UI
+
+## 0.5.14
+
+### Patch Changes
+
+- 1df8cfb: Split artifacts use case to document generator and code generator
+- 1b5a519: chore: improve dev experience with nodemon
+- b3eb0ba: Fix typing check issue
+- 556f33c: fix chromadb dependency issue
+- 2451539: fix: remove dead generated ai code
+- 7a70390: Deprecate pro mode
+
+## 0.5.13
+
+### Patch Changes
+
+- f4ca602: Add artifact use case for Typescript template
+- f4ca602: Update typescript use cases to use the new workflow engine
+
+## 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

packages/create-llama/CLAUDE.md

@@ -0,0 +1,108 @@
+# create-llama Package
+
+## Overview
+
+The `create-llama` package is a CLI tool for creating LlamaIndex-powered applications with one command. It's designed as a project generator that scaffolds various types of RAG (Retrieval-Augmented Generation) applications using different frameworks, databases, and AI model providers.
+
+## Package Structure
+
+### Core Files
+
+- **`index.ts`**: Main CLI entry point using Commander.js for argument parsing
+- **`create-app.ts`**: Core application creation logic and orchestration
+- **`package.json`**: Package configuration with binary entry point at `./dist/index.js`
+
+### Key Directories
+
+- **`helpers/`**: Utility functions for package management, file operations, and configuration
+- **`questions/`**: Interactive prompts for user configuration
+- **`templates/`**: Project templates for different frameworks and use cases
+- **`e2e/`**: End-to-end tests using Playwright
+
+## Core Functionality
+
+### CLI Interface
+
+The tool accepts numerous command-line options including:
+
+- Framework selection (`--framework`: nextjs, express, fastapi)
+- Template type (`--template`: streaming, multiagent, reflex, llamaindexserver)
+- Model providers (OpenAI, Anthropic, Groq, Ollama, etc.)
+- Vector databases (none, mongo, pg, pinecone, milvus, etc.)
+- Data sources (files, web URLs, databases)
+- Tools and observability options
+
+### Application Generation Flow
+
+1. **Project validation**: Checks project name validity and directory permissions
+2. **Interactive questioning**: Prompts user for configuration if not provided via CLI
+3. **Template installation**: Copies and configures appropriate templates
+4. **Environment setup**: Creates `.env` files with API keys and configuration
+5. **Dependencies**: Installs packages using detected/specified package manager
+6. **Post-install actions**: Can run the app, open VSCode, or install dependencies
+
+### Template System
+
+Templates are organized by:
+
+- **Framework**: NextJS (frontend), Express (Node backend), FastAPI (Python backend)
+- **Type**: Streaming chat, multiagent workflows, Reflex UI, LlamaIndex server
+- **Components**: Engines, loaders, providers, UI components, observability
+
+### Helper Functions
+
+Key helper modules include:
+
+- **Installation**: Package manager detection and dependency installation
+- **Data sources**: File copying, web scraping, database connection setup
+- **Providers**: Model provider configuration (OpenAI, Anthropic, etc.)
+- **Tools**: Integration with external tools (Wikipedia, weather, code generation)
+- **Environment**: `.env` file generation with API keys and settings
+
+## Development Commands
+
+### Build & Development
+
+- `npm run build`: Build the CLI using bash script
+- `npm run dev`: Watch mode development build
+- `npm run clean`: Clean build artifacts and temporary files
+
+### Testing
+
+- `npm run e2e`: Run all end-to-end tests
+- `npm run e2e:python`: Test Python-specific templates
+- `npm run e2e:typescript`: Test TypeScript-specific templates
+
+### Package Management
+
+- `npm run pack-install`: Create and install local package for testing
+
+## Architecture Notes
+
+### Model Configuration
+
+The tool supports multiple AI providers with a unified `ModelConfig` interface that includes:
+
+- Provider selection and API key management
+- Model and embedding model specification
+- Dimension configuration for embeddings
+
+### Data Source Handling
+
+Flexible data source configuration supporting:
+
+- Local files and directories
+- Web URLs with configurable crawling depth
+- Database connections with custom queries
+- Automatic file downloading and copying
+
+### Template Flexibility
+
+Templates use a component-based system allowing mix-and-match of:
+
+- Different frameworks (NextJS, Express, FastAPI)
+- Various vector databases
+- Multiple observability tools
+- Configurable tools and integrations
+
+This package serves as the foundation for rapidly prototyping and deploying LlamaIndex applications across different technology stacks and use cases.

packages/create-llama/create-app.ts

@@ -1,44 +1,34 @@
-/* eslint-disable import/no-extraneous-dependencies */
 import path from "path";
 import { green, yellow } from "picocolors";
 import { tryGitInit } from "./helpers/git";
 import { isFolderEmpty } from "./helpers/is-folder-empty";
-import { getOnline } from "./helpers/is-online";
 import { isWriteable } from "./helpers/is-writeable";
 import { makeDir } from "./helpers/make-dir";
 
 import terminalLink from "terminal-link";
-import type { InstallTemplateArgs, TemplateObservability } from "./helpers";
+import type { InstallTemplateArgs } from "./helpers";
 import { installTemplate } from "./helpers";
 import { templatesDir } from "./helpers/dir";
-import { toolsRequireConfig } from "./helpers/tools";
 import { configVSCode } from "./helpers/vscode";
 
 export type InstallAppArgs = Omit<
   InstallTemplateArgs,
-  "appName" | "root" | "isOnline" | "port"
+  "appName" | "root" | "port"
 > & {
   appPath: string;
-  frontend: boolean;
 };
 
 export async function createApp({
   template,
   framework,
-  ui,
   appPath,
   packageManager,
-  frontend,
   modelConfig,
   llamaCloudKey,
-  communityProjectConfig,
-  llamapack,
   vectorDb,
   postInstallAction,
   dataSources,
-  tools,
   useLlamaParse,
-  observability,
   useCase,
 }: InstallAppArgs): Promise<void> {
   const root = path.resolve(appPath);
@@ -60,9 +50,6 @@ export async function createApp({
     process.exit(1);
   }
 
-  const useYarn = packageManager === "yarn";
-  const isOnline = !useYarn || (await getOnline());
-
   console.log(`Creating a new LlamaIndex app in ${green(root)}.`);
   console.log();
 
@@ -71,36 +58,18 @@ export async function createApp({
     root,
     template,
     framework,
-    ui,
     packageManager,
-    isOnline,
     modelConfig,
     llamaCloudKey,
-    communityProjectConfig,
-    llamapack,
     vectorDb,
     postInstallAction,
     dataSources,
-    tools,
     useLlamaParse,
-    observability,
     useCase,
   };
 
   // Install backend
-  await installTemplate({ ...args, backend: true });
-
-  if (frontend && framework === "fastapi" && template !== "llamaindexserver") {
-    // install frontend
-    const frontendRoot = path.join(root, ".frontend");
-    await makeDir(frontendRoot);
-    await installTemplate({
-      ...args,
-      root: frontendRoot,
-      framework: "nextjs",
-      backend: false,
-    });
-  }
+  await installTemplate(args);
 
   await configVSCode(root, templatesDir, framework);
 
@@ -110,18 +79,6 @@ export async function createApp({
     console.log();
   }
 
-  if (toolsRequireConfig(tools) && template !== "llamaindexserver") {
-    const configFile =
-      framework === "fastapi" ? "config/tools.yaml" : "config/tools.json";
-    console.log(
-      yellow(
-        `You have selected tools that require configuration. Please configure them in the ${terminalLink(
-          configFile,
-          `file://${root}/${configFile}`,
-        )} file.`,
-      ),
-    );
-  }
   console.log("");
   console.log(`${green("Success!")} Created ${appName} at ${appPath}`);
 
@@ -132,8 +89,6 @@ export async function createApp({
     )} and learn how to get started.`,
   );
 
-  outputObservability(args.observability);
-
   if (
     dataSources.some((dataSource) => dataSource.type === "file") &&
     process.platform === "linux"
@@ -150,24 +105,3 @@ export async function createApp({
 
   console.log();
 }
-
-function outputObservability(observability?: TemplateObservability) {
-  switch (observability) {
-    case "traceloop":
-      console.log(
-        `\n${yellow("Observability")}: Visit the ${terminalLink(
-          "documentation",
-          "https://traceloop.com/docs/openllmetry/integrations",
-        )} to set up the environment variables and start seeing execution traces.`,
-      );
-      break;
-    case "llamatrace":
-      console.log(
-        `\n${yellow("Observability")}: LlamaTrace has been configured for your project. Visit the ${terminalLink(
-          "LlamaTrace dashboard",
-          "https://llamatrace.com/login",
-        )} to view your traces and monitor your application.`,
-      );
-      break;
-  }
-}

packages/create-llama/e2e/python/resolve_dependencies.spec.ts

@@ -0,0 +1,110 @@
+import { expect, test } from "@playwright/test";
+import { exec } from "child_process";
+import fs from "fs";
+import path from "path";
+import util from "util";
+import {
+  ALL_USE_CASES,
+  TemplateFramework,
+  TemplateVectorDB,
+} from "../../helpers/types";
+import { RunCreateLlamaOptions, createTestDir, runCreateLlama } from "../utils";
+
+const execAsync = util.promisify(exec);
+
+const templateFramework: TemplateFramework = "fastapi";
+const vectorDb: TemplateVectorDB = process.env.VECTORDB
+  ? (process.env.VECTORDB as TemplateVectorDB)
+  : "none";
+
+test.describe("Mypy check", () => {
+  test.describe.configure({ retries: 0 });
+
+  test.describe("LlamaIndexServer", async () => {
+    for (const useCase of ALL_USE_CASES) {
+      test(`should pass mypy for use case: ${useCase}`, async () => {
+        const cwd = await createTestDir();
+        await createAndCheckLlamaProject({
+          options: {
+            cwd,
+            templateFramework,
+            vectorDb,
+            port: 3000,
+            postInstallAction: "none",
+            llamaCloudProjectName: undefined,
+            llamaCloudIndexName: undefined,
+            useCase,
+          },
+        });
+      });
+    }
+  });
+});
+
+async function createAndCheckLlamaProject({
+  options,
+}: {
+  options: RunCreateLlamaOptions;
+}): Promise<{ pyprojectPath: string; projectPath: string }> {
+  const result = await runCreateLlama(options);
+  const name = result.projectName;
+  const projectPath = path.join(options.cwd, name);
+
+  // Check if the app folder exists
+  expect(fs.existsSync(projectPath)).toBeTruthy();
+
+  // Check if pyproject.toml exists
+  const pyprojectPath = path.join(projectPath, "pyproject.toml");
+  expect(fs.existsSync(pyprojectPath)).toBeTruthy();
+
+  // Modify environment for the command
+  const commandEnv = {
+    ...process.env,
+  };
+
+  console.log("Running uv venv...");
+  try {
+    const { stdout: venvStdout, stderr: venvStderr } = await execAsync(
+      "uv venv",
+      { cwd: projectPath, env: commandEnv },
+    );
+    console.log("uv venv stdout:", venvStdout);
+    console.error("uv venv stderr:", venvStderr);
+  } catch (error) {
+    console.error("Error running uv venv:", error);
+    throw error; // Re-throw error to fail the test
+  }
+
+  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 ....");
+  try {
+    const { stdout: mypyStdout, stderr: mypyStderr } = await execAsync(
+      "uv run mypy .",
+      { cwd: projectPath, env: commandEnv },
+    );
+    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
+  } catch (error) {
+    console.error("Error running mypy:", error);
+    throw error;
+  }
+
+  // If we reach this point without throwing an error, the test passes
+  expect(true).toBeTruthy();
+
+  return { pyprojectPath, projectPath };
+}

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

@@ -1,54 +1,44 @@
-/* eslint-disable turbo/no-undeclared-env-vars */
 import { expect, test } from "@playwright/test";
 import { ChildProcess } from "child_process";
 import fs from "fs";
 import path from "path";
-import type {
-  TemplateFramework,
-  TemplatePostInstallAction,
-  TemplateUI,
+import {
+  ALL_USE_CASES,
+  type TemplateFramework,
+  type TemplateVectorDB,
 } from "../../helpers";
-import { createTestDir, runCreateLlama, type AppType } from "../utils";
+import { createTestDir, runCreateLlama } from "../utils";
 
 const templateFramework: TemplateFramework = process.env.FRAMEWORK
   ? (process.env.FRAMEWORK as TemplateFramework)
   : "fastapi";
-const dataSource: string = "--example-file";
-const templateUI: TemplateUI = "shadcn";
-const templatePostInstallAction: TemplatePostInstallAction = "runApp";
-const appType: AppType = "--frontend";
-const userMessage = "Write a blog post about physical standards for letters";
-const templateUseCases = ["financial_report", "agentic_rag", "deep_research"];
+const vectorDb: TemplateVectorDB = process.env.VECTORDB
+  ? (process.env.VECTORDB as TemplateVectorDB)
+  : "none";
+const llamaCloudProjectName = "create-llama";
+const llamaCloudIndexName = "e2e-test";
 
-for (const useCase of templateUseCases) {
-  test.describe(`Test use case ${useCase} ${templateFramework} ${dataSource} ${templateUI} ${appType} ${templatePostInstallAction}`, async () => {
-    test.skip(
-      process.platform !== "linux" ||
-        process.env.DATASOURCE === "--no-files" ||
-        templateFramework === "express",
-      "The llamaindexserver template currently only works with nextjs, fastapi. We also only run on Linux to speed up tests.",
-    );
+const userMessage = "Write a blog post about physical standards for letters";
+
+for (const useCase of ALL_USE_CASES) {
+  test.describe(`Test use case ${useCase} ${templateFramework} ${vectorDb}`, async () => {
     let port: number;
     let cwd: string;
     let name: string;
     let appProcess: ChildProcess;
-    // Only test without using vector db for now
-    const vectorDb = "none";
 
     test.beforeAll(async () => {
       port = Math.floor(Math.random() * 10000) + 10000;
       cwd = await createTestDir();
       const result = await runCreateLlama({
         cwd,
-        templateType: "llamaindexserver",
         templateFramework,
-        dataSource,
         vectorDb,
         port,
-        postInstallAction: templatePostInstallAction,
-        templateUI,
-        appType,
+        postInstallAction: "runApp",
         useCase,
+        llamaCloudProjectName,
+        llamaCloudIndexName,
       });
       name = result.projectName;
       appProcess = result.appProcess;
@@ -60,22 +50,17 @@ for (const useCase of templateUseCases) {
     });
 
     test("Frontend should have a title", async ({ page }) => {
-      test.skip(
-        templatePostInstallAction !== "runApp" ||
-          templateFramework === "express",
-      );
       await page.goto(`http://localhost:${port}`);
-      await expect(page.getByText("Built by LlamaIndex")).toBeVisible();
+      await expect(page.getByText("Built by LlamaIndex")).toBeVisible({
+        timeout: 5 * 60 * 1000,
+      });
     });
 
     test("Frontend should be able to submit a message and receive the start of a streamed response", async ({
       page,
     }) => {
       test.skip(
-        templatePostInstallAction !== "runApp" ||
-          useCase === "financial_report" ||
-          useCase === "deep_research" ||
-          templateFramework === "express",
+        useCase === "financial_report" || useCase === "deep_research",
         "Skip chat tests for financial report and deep research.",
       );
       await page.goto(`http://localhost:${port}`);

packages/create-llama/e2e/typescript/eject.spec.ts

@@ -0,0 +1,70 @@
+import { expect, test } from "@playwright/test";
+import { ChildProcess, execSync } from "child_process";
+import fs from "fs";
+import path from "path";
+import { type TemplateFramework, type TemplateVectorDB } from "../../helpers";
+import { createTestDir, runCreateLlama } from "../utils";
+
+const templateFramework: TemplateFramework = "nextjs";
+const useCase = "code_generator";
+const vectorDb: TemplateVectorDB = process.env.VECTORDB
+  ? (process.env.VECTORDB as TemplateVectorDB)
+  : "none";
+
+const llamaCloudProjectName = "create-llama";
+const llamaCloudIndexName = "e2e-test";
+
+const ejectDir = "next";
+
+test.describe.skip(
+  `Test eject command for ${useCase} ${templateFramework} ${vectorDb}`,
+  async () => {
+    let port: number;
+    let cwd: string;
+    let name: string;
+    let appProcess: ChildProcess;
+
+    test.beforeAll(async () => {
+      port = Math.floor(Math.random() * 10000) + 10000;
+      cwd = await createTestDir();
+      const result = await runCreateLlama({
+        cwd,
+        templateFramework,
+        vectorDb,
+        port,
+        postInstallAction: "dependencies",
+        useCase,
+        llamaCloudProjectName,
+        llamaCloudIndexName,
+      });
+      name = result.projectName;
+      appProcess = result.appProcess;
+    });
+
+    test("Should successfully eject, install dependencies and build without errors", async ({
+      page,
+    }) => {
+      test.skip(
+        vectorDb === "llamacloud",
+        "Eject test only works with non-llamacloud",
+      );
+      // Run eject command
+      execSync("npm run eject", { cwd: path.join(cwd, name) });
+
+      // Verify next directory exists
+      const nextDirExists = fs.existsSync(path.join(cwd, name, ejectDir));
+      expect(nextDirExists).toBeTruthy();
+
+      // Install dependencies in next directory
+      execSync("npm install", { cwd: path.join(cwd, name, ejectDir) });
+
+      // Run build
+      execSync("npm run build", { cwd: path.join(cwd, name, ejectDir) });
+    });
+
+    // clean processes
+    test.afterAll(async () => {
+      appProcess?.kill();
+    });
+  },
+);

packages/create-llama/e2e/typescript/resolve_dependencies.spec.ts

@@ -0,0 +1,90 @@
+import { expect, test } from "@playwright/test";
+import { exec } from "child_process";
+import fs from "fs";
+import path from "path";
+import util from "util";
+import {
+  ALL_USE_CASES,
+  TemplateFramework,
+  TemplateUseCase,
+  TemplateVectorDB,
+} from "../../helpers/types";
+import { createTestDir, runCreateLlama } from "../utils";
+
+const execAsync = util.promisify(exec);
+
+const templateFramework: TemplateFramework = "nextjs";
+const vectorDb: TemplateVectorDB = process.env.VECTORDB
+  ? (process.env.VECTORDB as TemplateVectorDB)
+  : "none";
+
+test.describe("Test resolve TS dependencies", () => {
+  test.describe.configure({ retries: 0 });
+
+  for (const useCase of ALL_USE_CASES) {
+    const optionDescription = `useCase: ${useCase}, vectorDb: ${vectorDb}`;
+    test.describe(`${optionDescription}`, () => {
+      test(`${optionDescription}`, async () => {
+        await runTest({
+          useCase: useCase,
+          vectorDb: vectorDb,
+        });
+      });
+    });
+  }
+});
+
+async function runTest(options: {
+  useCase: TemplateUseCase;
+  vectorDb: TemplateVectorDB;
+}) {
+  const cwd = await createTestDir();
+
+  const result = await runCreateLlama({
+    cwd: cwd,
+    templateFramework: templateFramework,
+    vectorDb: options.vectorDb,
+    port: 3000,
+    postInstallAction: "none",
+    llamaCloudProjectName: undefined,
+    llamaCloudIndexName: undefined,
+    useCase: options.useCase,
+  });
+  const name = result.projectName;
+
+  // Check if the app folder exists
+  const appDir = path.join(cwd, name);
+  const dirExists = fs.existsSync(appDir);
+  expect(dirExists).toBeTruthy();
+
+  // Install dependencies using pnpm
+  try {
+    const { stderr: installStderr } = await execAsync(
+      "pnpm install --prefer-offline --ignore-workspace",
+      {
+        cwd: appDir,
+      },
+    );
+  } catch (error) {
+    console.error("Error installing dependencies:", error);
+    throw error;
+  }
+
+  // Run tsc type check and capture the output
+  try {
+    const { stdout, stderr } = await execAsync(
+      "pnpm exec tsc -b --diagnostics",
+      {
+        cwd: appDir,
+      },
+    );
+    // Check if there's any error output
+    expect(stderr).toBeFalsy();
+
+    // Log the stdout for debugging purposes
+    console.log("TypeScript type-check output:", stdout);
+  } catch (error) {
+    console.error("Error running tsc:", error);
+    throw error;
+  }
+}

packages/create-llama/e2e/utils.ts

@@ -6,13 +6,9 @@ import waitPort from "wait-port";
 import {
   TemplateFramework,
   TemplatePostInstallAction,
-  TemplateType,
-  TemplateUI,
   TemplateVectorDB,
 } from "../helpers";
 
-export type AppType = "--frontend" | "--no-frontend" | "";
-
 export type CreateLlamaResult = {
   projectName: string;
   appProcess: ChildProcess;
@@ -20,72 +16,36 @@ export type CreateLlamaResult = {
 
 export type RunCreateLlamaOptions = {
   cwd: string;
-  templateType: TemplateType;
   templateFramework: TemplateFramework;
-  dataSource: string;
   vectorDb: TemplateVectorDB;
   port: number;
   postInstallAction: TemplatePostInstallAction;
-  templateUI?: TemplateUI;
-  appType?: AppType;
+  useCase: string;
   llamaCloudProjectName?: string;
   llamaCloudIndexName?: string;
-  tools?: string;
-  useLlamaParse?: boolean;
-  observability?: string;
-  useCase?: string;
 };
 
 export async function runCreateLlama({
   cwd,
-  templateType,
   templateFramework,
-  dataSource,
   vectorDb,
   port,
   postInstallAction,
-  templateUI,
-  appType,
+  useCase,
   llamaCloudProjectName,
   llamaCloudIndexName,
-  tools,
-  useLlamaParse,
-  observability,
-  useCase,
 }: RunCreateLlamaOptions): Promise<CreateLlamaResult> {
   if (!process.env.OPENAI_API_KEY || !process.env.LLAMA_CLOUD_API_KEY) {
     throw new Error(
       "Setting the OPENAI_API_KEY and LLAMA_CLOUD_API_KEY is mandatory to run tests",
     );
   }
-  const name = [
-    templateType,
-    templateFramework,
-    dataSource.split(" ")[0],
-    templateUI,
-    appType,
-  ].join("-");
-
-  // Handle different data source types
-  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")) {
-    const dbSource = dataSource.split(" ")[1];
-    dataSourceArgs.push("--db-source", dbSource);
-  } else {
-    dataSourceArgs.push(dataSource);
-  }
-
+  const name = [templateFramework, useCase, vectorDb].join("-");
   const commandArgs = [
     "create-llama",
     name,
-    "--template",
-    templateType,
     "--framework",
     templateFramework,
-    ...dataSourceArgs,
     "--vector-db",
     vectorDb,
     "--use-npm",
@@ -93,35 +53,10 @@ export async function runCreateLlama({
     port,
     "--post-install-action",
     postInstallAction,
-    "--tools",
-    tools ?? "none",
-    "--observability",
-    "none",
+    "--use-case",
+    useCase,
   ];
 
-  if (templateUI) {
-    commandArgs.push("--ui", templateUI);
-  }
-  if (appType) {
-    commandArgs.push(appType);
-  }
-  if (useLlamaParse) {
-    commandArgs.push("--use-llama-parse");
-  } else {
-    commandArgs.push("--no-llama-parse");
-  }
-  if (observability) {
-    commandArgs.push("--observability", observability);
-  }
-  if (
-    (templateType === "multiagent" ||
-      templateType === "reflex" ||
-      templateType === "llamaindexserver") &&
-    useCase
-  ) {
-    commandArgs.push("--use-case", useCase);
-  }
-
   const command = commandArgs.join(" ");
   console.log(`running command '${command}' in ${cwd}`);
   const appProcess = exec(command, {

packages/create-llama/helpers/copy.ts

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

packages/create-llama/helpers/datasources.ts

@@ -0,0 +1,51 @@
+import path from "path";
+import { templatesDir } from "./dir";
+import { TemplateDataSource } from "./types";
+
+export const EXAMPLE_FILE: TemplateDataSource = {
+  type: "file",
+  config: {
+    path: path.join(templatesDir, "components", "data", "101.pdf"),
+  },
+};
+
+export const EXAMPLE_10K_SEC_FILES: TemplateDataSource[] = [
+  {
+    type: "file",
+    config: {
+      url: new URL(
+        "https://s2.q4cdn.com/470004039/files/doc_earnings/2023/q4/filing/_10-K-Q4-2023-As-Filed.pdf",
+      ),
+      filename: "apple_10k_report.pdf",
+    },
+  },
+  {
+    type: "file",
+    config: {
+      url: new URL(
+        "https://ir.tesla.com/_flysystem/s3/sec/000162828024002390/tsla-20231231-gen.pdf",
+      ),
+      filename: "tesla_10k_report.pdf",
+    },
+  },
+];
+
+export const EXAMPLE_GDPR: TemplateDataSource = {
+  type: "file",
+  config: {
+    url: new URL(
+      "https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32016R0679",
+    ),
+    filename: "gdpr.pdf",
+  },
+};
+
+export const AI_REPORTS: TemplateDataSource = {
+  type: "file",
+  config: {
+    url: new URL(
+      "https://www.europarl.europa.eu/RegData/etudes/ATAG/2024/760392/EPRS_ATA(2024)760392_EN.pdf",
+    ),
+    filename: "EPRS_ATA_2024_760392_EN.pdf",
+  },
+};

packages/create-llama/helpers/env-variables.ts

@@ -1,24 +1,15 @@
 import fs from "fs/promises";
 import path from "path";
-import { TOOL_SYSTEM_PROMPT_ENV_VAR, Tool } from "./tools";
 import {
   InstallTemplateArgs,
   ModelConfig,
-  TemplateDataSource,
   TemplateFramework,
-  TemplateObservability,
   TemplateType,
   TemplateVectorDB,
 } from "./types";
 
 import { TSYSTEMS_LLMHUB_API_URL } from "./providers/llmhub";
 
-const DEFAULT_SYSTEM_PROMPT =
-  "You are a helpful assistant who helps users with their questions.";
-
-const DATA_SOURCES_PROMPT =
-  "You have access to a knowledge base including the facts that you should start with to find the answer for the user question. Use the query engine tool to retrieve the facts from the knowledge base.";
-
 export type EnvVar = {
   name?: string;
   description?: string;
@@ -181,7 +172,7 @@ const getVectorDBEnvs = (
             ]
           : []),
       ];
-    case "chroma":
+    case "chroma": {
       const envs = [
         {
           name: "CHROMA_COLLECTION",
@@ -206,6 +197,7 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
         });
       }
       return envs;
+    }
     case "weaviate":
       return [
         {
@@ -238,11 +230,6 @@ Otherwise, use CHROMA_HOST and CHROMA_PORT config above`,
 
 const getModelEnvs = (modelConfig: ModelConfig): EnvVar[] => {
   return [
-    {
-      name: "MODEL_PROVIDER",
-      description: "The provider for the AI models to use.",
-      value: modelConfig.provider,
-    },
     {
       name: "MODEL",
       description: "The name of LLM model to use.",
@@ -253,11 +240,6 @@ const getModelEnvs = (modelConfig: ModelConfig): EnvVar[] => {
       description: "Name of the embedding model to use.",
       value: modelConfig.embeddingModel,
     },
-    {
-      name: "EMBEDDING_DIM",
-      description: "Dimension of the embedding model to use.",
-      value: modelConfig.dimensions.toString(),
-    },
     {
       name: "CONVERSATION_STARTERS",
       description: "The questions to help users get started (multi-line).",
@@ -385,25 +367,10 @@ 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[] = [];
   if (framework === "fastapi") {
     result.push(
       ...[
@@ -420,149 +387,10 @@ const getFrameworkEnvs = (
       ],
     );
   }
-  if (framework === "nextjs" && template !== "llamaindexserver") {
-    result.push({
-      name: "NEXT_PUBLIC_CHAT_API",
-      description:
-        "The API for the chat endpoint. Set when using a custom backend (e.g. Express). Use full URL like http://localhost:8000/api/chat",
-    });
-  }
+
   return result;
 };
 
-const getEngineEnvs = (): EnvVar[] => {
-  return [
-    {
-      name: "TOP_K",
-      description:
-        "The number of similar embeddings to return when retrieving documents.",
-    },
-  ];
-};
-
-const getToolEnvs = (tools?: Tool[]): EnvVar[] => {
-  if (!tools?.length) return [];
-  const toolEnvs: EnvVar[] = [];
-  tools.forEach((tool) => {
-    if (tool.envVars?.length) {
-      toolEnvs.push(
-        // Don't include the system prompt env var here
-        // It should be handled separately by merging with the default system prompt
-        ...tool.envVars.filter(
-          (env) => env.name !== TOOL_SYSTEM_PROMPT_ENV_VAR,
-        ),
-      );
-    }
-  });
-  return toolEnvs;
-};
-
-const getSystemPromptEnv = (
-  tools?: Tool[],
-  dataSources?: TemplateDataSource[],
-  template?: TemplateType,
-): EnvVar[] => {
-  const systemPromptEnv: EnvVar[] = [];
-  // build tool system prompt by merging all tool system prompts
-  // multiagent template doesn't need system prompt
-  if (template !== "multiagent") {
-    let toolSystemPrompt = "";
-    tools?.forEach((tool) => {
-      const toolSystemPromptEnv = tool.envVars?.find(
-        (env) => env.name === TOOL_SYSTEM_PROMPT_ENV_VAR,
-      );
-      if (toolSystemPromptEnv) {
-        toolSystemPrompt += toolSystemPromptEnv.value + "\n";
-      }
-    });
-
-    const systemPrompt =
-      '"' +
-      DEFAULT_SYSTEM_PROMPT +
-      (dataSources?.length ? `\n${DATA_SOURCES_PROMPT}` : "") +
-      (toolSystemPrompt ? `\n${toolSystemPrompt}` : "") +
-      '"';
-
-    systemPromptEnv.push({
-      name: "SYSTEM_PROMPT",
-      description: "The system prompt for the AI model.",
-      value: systemPrompt,
-    });
-  }
-  if (tools?.length == 0 && (dataSources?.length ?? 0 > 0)) {
-    const citationPrompt = `'You have provided information from a knowledge base that has been passed to you in nodes of information.
-Each node has useful metadata such as node ID, file name, page, etc.
-Please add the citation to the data node for each sentence or paragraph that you reference in the provided information.
-The citation format is: . [citation:<node_id>]()
-Where the <node_id> is the unique identifier of the data node.
-
-Example:
-We have two nodes:
-  node_id: xyz
-  file_name: llama.pdf
-  
-  node_id: abc
-  file_name: animal.pdf
-
-User question: Tell me a fun fact about Llama.
-Your answer:
-A baby llama is called "Cria" [citation:xyz]().
-It often live in desert [citation:abc]().
-It\\'s cute animal.
-'`;
-    systemPromptEnv.push({
-      name: "SYSTEM_CITATION_PROMPT",
-      description:
-        "An additional system prompt to add citation when responding to user questions.",
-      value: citationPrompt,
-    });
-  }
-
-  return systemPromptEnv;
-};
-
-const getTemplateEnvs = (template?: TemplateType): EnvVar[] => {
-  const nextQuestionEnvs: EnvVar[] = [
-    {
-      name: "NEXT_QUESTION_PROMPT",
-      description: `Customize prompt to generate the next question suggestions based on the conversation history.
-Disable this prompt to disable the next question suggestions feature.`,
-      value: `"You're a helpful assistant! Your task is to suggest the next question that user might ask. 
-Here is the conversation history
----------------------
-{conversation}
----------------------
-Given the conversation history, please give me 3 questions that user might ask next!
-Your answer should be wrapped in three sticks which follows the following format:
-\`\`\`
-<question 1>
-<question 2>
-<question 3>
-\`\`\`"`,
-    },
-  ];
-
-  if (template === "multiagent" || template === "streaming") {
-    return nextQuestionEnvs;
-  }
-  return [];
-};
-
-const getObservabilityEnvs = (
-  observability?: TemplateObservability,
-): EnvVar[] => {
-  if (observability === "llamatrace") {
-    return [
-      {
-        name: "PHOENIX_API_KEY",
-        description:
-          "API key for LlamaTrace observability. Retrieve from https://llamatrace.com/login",
-      },
-    ];
-  }
-  return [];
-};
-
 export const createBackendEnvFile = async (
   root: string,
   opts: Pick<
@@ -574,8 +402,6 @@ export const createBackendEnvFile = async (
     | "dataSources"
     | "template"
     | "port"
-    | "tools"
-    | "observability"
     | "useLlamaParse"
   >,
 ) => {
@@ -592,45 +418,11 @@ export const createBackendEnvFile = async (
         ]
       : []),
     ...getVectorDBEnvs(opts.vectorDb, opts.framework, opts.template),
-    ...getToolEnvs(opts.tools),
-    ...getFrameworkEnvs(opts.framework, opts.template, opts.port),
-    // 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),
-        ]),
+    ...getFrameworkEnvs(opts.framework, opts.port),
+    ...getModelEnvs(opts.modelConfig),
   ];
   // Render and write env file
   const content = renderEnvVar(envVars);
   await fs.writeFile(path.join(root, envFileName), content);
   console.log(`Created '${envFileName}' file. Please check the settings.`);
 };
-
-export const createFrontendEnvFile = async (
-  root: string,
-  opts: {
-    vectorDb?: TemplateVectorDB;
-  },
-) => {
-  const defaultFrontendEnvs = [
-    {
-      name: "NEXT_PUBLIC_USE_LLAMACLOUD",
-      description: "Let's the user change indexes in LlamaCloud projects",
-      value: opts.vectorDb === "llamacloud" ? "true" : "false",
-    },
-  ];
-  const content = renderEnvVar(defaultFrontendEnvs);
-  await fs.writeFile(path.join(root, ".env"), content);
-};

packages/create-llama/helpers/git.ts

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

packages/create-llama/helpers/index.ts

@@ -4,15 +4,10 @@ import path from "path";
 import picocolors, { cyan } from "picocolors";
 
 import fsExtra from "fs-extra";
-import { writeLoadersConfig } from "./datasources";
-import { createBackendEnvFile, createFrontendEnvFile } from "./env-variables";
+import { createBackendEnvFile } 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";
 import {
   FileSourceConfig,
   InstallTemplateArgs,
@@ -22,6 +17,7 @@ import {
   TemplateVectorDB,
 } from "./types";
 import { installTSTemplate } from "./typescript";
+import { isHavingUvLockFile, tryUvRun } from "./uv";
 
 const checkForGenerateScript = (
   modelConfig: ModelConfig,
@@ -56,6 +52,7 @@ const checkForGenerateScript = (
 async function generateContextData(
   framework: TemplateFramework,
   modelConfig: ModelConfig,
+  dataSources: TemplateDataSource[],
   packageManager?: PackageManager,
   vectorDb?: TemplateVectorDB,
   llamaCloudKey?: string,
@@ -64,7 +61,7 @@ async function generateContextData(
   if (packageManager) {
     const runGenerate = `${cyan(
       framework === "fastapi"
-        ? "poetry run generate"
+        ? "uv run generate"
         : `${packageManager} run generate`,
     )}`;
 
@@ -78,19 +75,30 @@ async function generateContextData(
     if (!missingSettings.length) {
       // If all the required environment variables are set, run the generate script
       if (framework === "fastapi") {
-        if (isHavingPoetryLockFile()) {
+        if (isHavingUvLockFile()) {
           console.log(`Running ${runGenerate} to generate the context data.`);
-          const result = tryPoetryRun("poetry run generate");
+          const result = tryUvRun("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.`);
-        await callPackageManager(packageManager, true, ["run", "generate"]);
+
+        const shouldRunGenerate = dataSources.length > 0;
+
+        if (shouldRunGenerate) {
+          await callPackageManager(packageManager, true, ["run", "generate"]);
+        }
         return;
       }
     }
@@ -103,7 +111,7 @@ async function generateContextData(
 const downloadFile = async (url: string, destPath: string) => {
   const response = await fetch(url);
   const fileBuffer = await response.arrayBuffer();
-  await fsExtra.writeFile(destPath, Buffer.from(fileBuffer));
+  await fsExtra.writeFile(destPath, new Uint8Array(fileBuffer));
 };
 
 const prepareContextData = async (
@@ -139,98 +147,46 @@ const prepareContextData = async (
   }
 };
 
-const installCommunityProject = async ({
-  root,
-  communityProjectConfig,
-}: Pick<InstallTemplateArgs, "root" | "communityProjectConfig">) => {
-  const { owner, repo, branch, filePath } = communityProjectConfig!;
-  console.log("\nInstalling community project:", filePath || repo);
-  await downloadAndExtractRepo(root, {
-    username: owner,
-    name: repo,
-    branch,
-    filePath: filePath || "",
-  });
-};
-
-export const installTemplate = async (
-  props: InstallTemplateArgs & { backend: boolean },
-) => {
+export const installTemplate = async (props: InstallTemplateArgs) => {
   process.chdir(props.root);
 
-  if (props.template === "community" && props.communityProjectConfig) {
-    await installCommunityProject(props);
-    return;
-  }
-
-  if (props.template === "llamapack" && props.llamapack) {
-    await installLlamapackProject(props);
-    return;
-  }
-
   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,
+  // 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.
+  await createBackendEnvFile(props.root, props);
+
+  await prepareContextData(
+    props.root,
+    props.dataSources.filter((ds) => ds.type === "file"),
+  );
+
+  if (
+    props.dataSources.length > 0 &&
+    (props.postInstallAction === "runApp" ||
+      props.postInstallAction === "dependencies")
+  ) {
+    console.log("\nGenerating context data...\n");
+    await generateContextData(
+      props.framework,
+      props.modelConfig,
+      props.dataSources,
+      props.packageManager,
+      props.vectorDb,
+      props.llamaCloudKey,
+      props.useLlamaParse,
     );
-    if (props.vectorDb !== "llamacloud") {
-      // write loaders configuration (currently Python only)
-      // not needed for LlamaCloud as it has its own loaders
-      await writeLoadersConfig(
-        props.root,
-        props.dataSources,
-        props.useLlamaParse,
-      );
-    }
   }
 
-  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") {
-      await createBackendEnvFile(props.root, props);
-    }
-
-    await prepareContextData(
-      props.root,
-      props.dataSources.filter((ds) => ds.type === "file"),
-    );
-
-    if (
-      props.dataSources.length > 0 &&
-      (props.postInstallAction === "runApp" ||
-        props.postInstallAction === "dependencies")
-    ) {
-      console.log("\nGenerating context data...\n");
-      await generateContextData(
-        props.framework,
-        props.modelConfig,
-        props.packageManager,
-        props.vectorDb,
-        props.llamaCloudKey,
-        props.useLlamaParse,
-      );
-    }
-
-    // Create outputs directory
-    await makeDir(path.join(props.root, "output/tools"));
-    await makeDir(path.join(props.root, "output/uploaded"));
-    await makeDir(path.join(props.root, "output/llamacloud"));
-  } else {
-    // this is a frontend for a full-stack app, create .env file with model information
-    await createFrontendEnvFile(props.root, {
-      vectorDb: props.vectorDb,
-    });
-  }
+  // Create outputs directory
+  await makeDir(path.join(props.root, "output/tools"));
+  await makeDir(path.join(props.root, "output/uploaded"));
+  await makeDir(path.join(props.root, "output/llamacloud"));
 };
 
 export * from "./types";

packages/create-llama/helpers/install.ts

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

packages/create-llama/helpers/is-folder-empty.ts

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

packages/create-llama/helpers/models.ts

@@ -0,0 +1,12 @@
+import { ModelConfig } from "./types";
+
+export const getGpt41ModelConfig = (): ModelConfig => ({
+  provider: "openai",
+  apiKey: process.env.OPENAI_API_KEY,
+  model: "gpt-4.1",
+  embeddingModel: "text-embedding-3-large",
+  dimensions: 1536,
+  isConfigured(): boolean {
+    return !!process.env.OPENAI_API_KEY;
+  },
+});

packages/create-llama/helpers/providers/anthropic.ts

@@ -31,17 +31,9 @@ const EMBEDDING_MODELS: Record<HuggingFaceEmbeddingModelType, ModelData> = {
 const DEFAULT_EMBEDDING_MODEL = Object.keys(EMBEDDING_MODELS)[0];
 const DEFAULT_DIMENSIONS = Object.values(EMBEDDING_MODELS)[0].dimensions;
 
-type AnthropicQuestionsParams = {
-  apiKey?: string;
-  askModels: boolean;
-};
-
-export async function askAnthropicQuestions({
-  askModels,
-  apiKey,
-}: AnthropicQuestionsParams): Promise<ModelConfigParams> {
+export async function askAnthropicQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey,
+    apiKey: process.env.ANTHROPIC_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: DEFAULT_DIMENSIONS,
@@ -69,35 +61,33 @@ export async function askAnthropicQuestions({
     config.apiKey = key || process.env.ANTHROPIC_API_KEY;
   }
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: MODELS.map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: MODELS.map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions =
-      EMBEDDING_MODELS[
-        embeddingModel as HuggingFaceEmbeddingModelType
-      ].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions =
+    EMBEDDING_MODELS[
+      embeddingModel as HuggingFaceEmbeddingModelType
+    ].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/azure.ts

@@ -1,5 +1,5 @@
 import prompts from "prompts";
-import { ModelConfigParams, ModelConfigQuestionsParams } from ".";
+import { ModelConfigParams } from ".";
 import { questionHandlers } from "../../questions/utils";
 
 const ALL_AZURE_OPENAI_CHAT_MODELS: Record<string, { openAIModel: string }> = {
@@ -51,12 +51,9 @@ const ALL_AZURE_OPENAI_EMBEDDING_MODELS: Record<
 const DEFAULT_MODEL = "gpt-4o";
 const DEFAULT_EMBEDDING_MODEL = "text-embedding-3-large";
 
-export async function askAzureQuestions({
-  openAiKey,
-  askModels,
-}: ModelConfigQuestionsParams): Promise<ModelConfigParams> {
+export async function askAzureQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey: openAiKey || process.env.AZURE_OPENAI_KEY,
+    apiKey: process.env.AZURE_OPENAI_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: getDimensions(DEFAULT_EMBEDDING_MODEL),
@@ -66,32 +63,30 @@ export async function askAzureQuestions({
     },
   };
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: getAvailableModelChoices(),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: getAvailableModelChoices(),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: getAvailableEmbeddingModelChoices(),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = getDimensions(embeddingModel);
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: getAvailableEmbeddingModelChoices(),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = getDimensions(embeddingModel);
 
   return config;
 }

packages/create-llama/helpers/providers/gemini.ts

@@ -2,7 +2,15 @@ import prompts from "prompts";
 import { ModelConfigParams } from ".";
 import { questionHandlers, toChoice } from "../../questions/utils";
 
-const MODELS = ["gemini-1.5-pro-latest", "gemini-pro", "gemini-pro-vision"];
+const MODELS = [
+  "gemini-2.5-pro",
+  "gemini-2.5-flash",
+  "gemini-2.0-flash",
+  "gemini-2.0-flash-lite",
+  "gemini-1.5-pro-latest",
+  "gemini-pro",
+  "gemini-pro-vision",
+];
 type ModelData = {
   dimensions: number;
 };
@@ -15,17 +23,9 @@ const DEFAULT_MODEL = MODELS[0];
 const DEFAULT_EMBEDDING_MODEL = Object.keys(EMBEDDING_MODELS)[0];
 const DEFAULT_DIMENSIONS = Object.values(EMBEDDING_MODELS)[0].dimensions;
 
-type GeminiQuestionsParams = {
-  apiKey?: string;
-  askModels: boolean;
-};
-
-export async function askGeminiQuestions({
-  askModels,
-  apiKey,
-}: GeminiQuestionsParams): Promise<ModelConfigParams> {
+export async function askGeminiQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey,
+    apiKey: process.env.GOOGLE_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: DEFAULT_DIMENSIONS,
@@ -53,32 +53,30 @@ export async function askGeminiQuestions({
     config.apiKey = key || process.env.GOOGLE_API_KEY;
   }
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: MODELS.map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: MODELS.map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/groq.ts

@@ -71,17 +71,9 @@ const EMBEDDING_MODELS: Record<HuggingFaceEmbeddingModelType, ModelData> = {
 const DEFAULT_EMBEDDING_MODEL = Object.keys(EMBEDDING_MODELS)[0];
 const DEFAULT_DIMENSIONS = Object.values(EMBEDDING_MODELS)[0].dimensions;
 
-type GroqQuestionsParams = {
-  apiKey?: string;
-  askModels: boolean;
-};
-
-export async function askGroqQuestions({
-  askModels,
-  apiKey,
-}: GroqQuestionsParams): Promise<ModelConfigParams> {
+export async function askGroqQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey,
+    apiKey: process.env.GROQ_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: DEFAULT_DIMENSIONS,
@@ -109,37 +101,35 @@ export async function askGroqQuestions({
     config.apiKey = key || process.env.GROQ_API_KEY;
   }
 
-  if (askModels) {
-    const modelChoices = await getAvailableModelChoicesGroq(config.apiKey!);
+  const modelChoices = await getAvailableModelChoicesGroq(config.apiKey!);
 
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: modelChoices,
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: modelChoices,
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions =
-      EMBEDDING_MODELS[
-        embeddingModel as HuggingFaceEmbeddingModelType
-      ].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions =
+    EMBEDDING_MODELS[
+      embeddingModel as HuggingFaceEmbeddingModelType
+    ].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/huggingface.ts

@@ -21,13 +21,7 @@ const DEFAULT_MODEL = MODELS[0];
 const DEFAULT_EMBEDDING_MODEL = Object.keys(EMBEDDING_MODELS)[0];
 const DEFAULT_DIMENSIONS = Object.values(EMBEDDING_MODELS)[0].dimensions;
 
-type HuggingfaceQuestionsParams = {
-  askModels: boolean;
-};
-
-export async function askHuggingfaceQuestions({
-  askModels,
-}: HuggingfaceQuestionsParams): Promise<ModelConfigParams> {
+export async function askHuggingfaceQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
@@ -37,32 +31,30 @@ export async function askHuggingfaceQuestions({
     },
   };
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which Hugging Face model would you like to use?",
-        choices: MODELS.map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which Hugging Face model would you like to use?",
+      choices: MODELS.map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/index.ts

@@ -0,0 +1,81 @@
+import prompts from "prompts";
+import { questionHandlers } from "../../questions/utils";
+import { ModelConfig, TemplateFramework } from "../types";
+import { askAnthropicQuestions } from "./anthropic";
+import { askAzureQuestions } from "./azure";
+import { askGeminiQuestions } from "./gemini";
+import { askGroqQuestions } from "./groq";
+import { askHuggingfaceQuestions } from "./huggingface";
+import { askLLMHubQuestions } from "./llmhub";
+import { askMistralQuestions } from "./mistral";
+import { askOllamaQuestions } from "./ollama";
+import { askOpenAIQuestions } from "./openai";
+
+export type ModelConfigQuestionsParams = {
+  framework?: TemplateFramework;
+};
+
+export type ModelConfigParams = Omit<ModelConfig, "provider">;
+
+export async function askModelConfig({
+  framework,
+}: ModelConfigQuestionsParams): Promise<ModelConfig> {
+  const choices = [
+    { title: "OpenAI", value: "openai" },
+    { title: "Groq", value: "groq" },
+    { title: "Ollama", value: "ollama" },
+    { title: "Anthropic", value: "anthropic" },
+    { title: "Gemini", value: "gemini" },
+    { title: "Mistral", value: "mistral" },
+    { title: "AzureOpenAI", value: "azure-openai" },
+  ];
+
+  if (framework === "fastapi") {
+    choices.push({ title: "T-Systems", value: "t-systems" });
+    choices.push({ title: "Huggingface", value: "huggingface" });
+  }
+  const { provider: modelProvider } = await prompts(
+    {
+      type: "select",
+      name: "provider",
+      message: "Which model provider would you like to use",
+      choices: choices,
+      initial: 0,
+    },
+    questionHandlers,
+  );
+
+  let modelConfig: ModelConfigParams;
+  switch (modelProvider) {
+    case "ollama":
+      modelConfig = await askOllamaQuestions();
+      break;
+    case "groq":
+      modelConfig = await askGroqQuestions();
+      break;
+    case "anthropic":
+      modelConfig = await askAnthropicQuestions();
+      break;
+    case "gemini":
+      modelConfig = await askGeminiQuestions();
+      break;
+    case "mistral":
+      modelConfig = await askMistralQuestions();
+      break;
+    case "azure-openai":
+      modelConfig = await askAzureQuestions();
+      break;
+    case "t-systems":
+      modelConfig = await askLLMHubQuestions();
+      break;
+    case "huggingface":
+      modelConfig = await askHuggingfaceQuestions();
+      break;
+    default:
+      modelConfig = await askOpenAIQuestions();
+  }
+  return {
+    ...modelConfig,
+    provider: modelProvider,
+  };
+}

packages/create-llama/helpers/providers/llmhub.ts

@@ -31,17 +31,9 @@ const LLMHUB_EMBEDDING_MODELS = [
   "text-embedding-bge-m3",
 ];
 
-type LLMHubQuestionsParams = {
-  apiKey?: string;
-  askModels: boolean;
-};
-
-export async function askLLMHubQuestions({
-  askModels,
-  apiKey,
-}: LLMHubQuestionsParams): Promise<ModelConfigParams> {
+export async function askLLMHubQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey,
+    apiKey: process.env.T_SYSTEMS_LLMHUB_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: getDimensions(DEFAULT_EMBEDDING_MODEL),
@@ -61,11 +53,10 @@ export async function askLLMHubQuestions({
       {
         type: "text",
         name: "key",
-        message: askModels
-          ? "Please provide your LLMHub API key (or leave blank to use T_SYSTEMS_LLMHUB_API_KEY env variable):"
-          : "Please provide your LLMHub API key (leave blank to skip):",
+        message:
+          "Please provide your LLMHub API key (or leave blank to use T_SYSTEMS_LLMHUB_API_KEY env variable):",
         validate: (value: string) => {
-          if (askModels && !value) {
+          if (!value) {
             if (process.env.T_SYSTEMS_LLMHUB_API_KEY) {
               return true;
             }
@@ -79,32 +70,30 @@ export async function askLLMHubQuestions({
     config.apiKey = key || process.env.T_SYSTEMS_LLMHUB_API_KEY;
   }
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: await getAvailableModelChoices(false, config.apiKey),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: await getAvailableModelChoices(false, config.apiKey),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: await getAvailableModelChoices(true, config.apiKey),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = getDimensions(embeddingModel);
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: await getAvailableModelChoices(true, config.apiKey),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = getDimensions(embeddingModel);
 
   return config;
 }

packages/create-llama/helpers/providers/mistral.ts

@@ -14,17 +14,9 @@ const DEFAULT_MODEL = MODELS[0];
 const DEFAULT_EMBEDDING_MODEL = Object.keys(EMBEDDING_MODELS)[0];
 const DEFAULT_DIMENSIONS = Object.values(EMBEDDING_MODELS)[0].dimensions;
 
-type MistralQuestionsParams = {
-  apiKey?: string;
-  askModels: boolean;
-};
-
-export async function askMistralQuestions({
-  askModels,
-  apiKey,
-}: MistralQuestionsParams): Promise<ModelConfigParams> {
+export async function askMistralQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey,
+    apiKey: process.env.MISTRAL_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: DEFAULT_DIMENSIONS,
@@ -52,32 +44,30 @@ export async function askMistralQuestions({
     config.apiKey = key || process.env.MISTRAL_API_KEY;
   }
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: MODELS.map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: MODELS.map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/ollama.ts

@@ -17,13 +17,7 @@ const EMBEDDING_MODELS: Record<string, ModelData> = {
 };
 const DEFAULT_EMBEDDING_MODEL: string = Object.keys(EMBEDDING_MODELS)[0];
 
-type OllamaQuestionsParams = {
-  askModels: boolean;
-};
-
-export async function askOllamaQuestions({
-  askModels,
-}: OllamaQuestionsParams): Promise<ModelConfigParams> {
+export async function askOllamaQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
@@ -33,34 +27,32 @@ export async function askOllamaQuestions({
     },
   };
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: MODELS.map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    await ensureModel(model);
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: MODELS.map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  await ensureModel(model);
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    await ensureModel(embeddingModel);
-    config.embeddingModel = embeddingModel;
-    config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: Object.keys(EMBEDDING_MODELS).map(toChoice),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  await ensureModel(embeddingModel);
+  config.embeddingModel = embeddingModel;
+  config.dimensions = EMBEDDING_MODELS[embeddingModel].dimensions;
 
   return config;
 }

packages/create-llama/helpers/providers/openai.ts

@@ -2,8 +2,7 @@ import got from "got";
 import ora from "ora";
 import { red } from "picocolors";
 import prompts from "prompts";
-import { ModelConfigParams, ModelConfigQuestionsParams } from ".";
-import { isCI } from "../../questions";
+import { ModelConfigParams } from ".";
 import { questionHandlers } from "../../questions/utils";
 
 const OPENAI_API_URL = "https://api.openai.com/v1";
@@ -11,12 +10,9 @@ const OPENAI_API_URL = "https://api.openai.com/v1";
 const DEFAULT_MODEL = "gpt-4o-mini";
 const DEFAULT_EMBEDDING_MODEL = "text-embedding-3-large";
 
-export async function askOpenAIQuestions({
-  openAiKey,
-  askModels,
-}: ModelConfigQuestionsParams): Promise<ModelConfigParams> {
+export async function askOpenAIQuestions(): Promise<ModelConfigParams> {
   const config: ModelConfigParams = {
-    apiKey: openAiKey,
+    apiKey: process.env.OPENAI_API_KEY,
     model: DEFAULT_MODEL,
     embeddingModel: DEFAULT_EMBEDDING_MODEL,
     dimensions: getDimensions(DEFAULT_EMBEDDING_MODEL),
@@ -31,16 +27,15 @@ export async function askOpenAIQuestions({
     },
   };
 
-  if (!config.apiKey && !isCI) {
+  if (!config.apiKey) {
     const { key } = await prompts(
       {
         type: "text",
         name: "key",
-        message: askModels
-          ? "Please provide your OpenAI API key (or leave blank to use OPENAI_API_KEY env variable):"
-          : "Please provide your OpenAI API key (leave blank to skip):",
+        message:
+          "Please provide your OpenAI API key (or leave blank to use OPENAI_API_KEY env variable):",
         validate: (value: string) => {
-          if (askModels && !value) {
+          if (!value) {
             if (process.env.OPENAI_API_KEY) {
               return true;
             }
@@ -54,32 +49,30 @@ export async function askOpenAIQuestions({
     config.apiKey = key || process.env.OPENAI_API_KEY;
   }
 
-  if (askModels) {
-    const { model } = await prompts(
-      {
-        type: "select",
-        name: "model",
-        message: "Which LLM model would you like to use?",
-        choices: await getAvailableModelChoices(false, config.apiKey),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.model = model;
+  const { model } = await prompts(
+    {
+      type: "select",
+      name: "model",
+      message: "Which LLM model would you like to use?",
+      choices: await getAvailableModelChoices(false, config.apiKey),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.model = model;
 
-    const { embeddingModel } = await prompts(
-      {
-        type: "select",
-        name: "embeddingModel",
-        message: "Which embedding model would you like to use?",
-        choices: await getAvailableModelChoices(true, config.apiKey),
-        initial: 0,
-      },
-      questionHandlers,
-    );
-    config.embeddingModel = embeddingModel;
-    config.dimensions = getDimensions(embeddingModel);
-  }
+  const { embeddingModel } = await prompts(
+    {
+      type: "select",
+      name: "embeddingModel",
+      message: "Which embedding model would you like to use?",
+      choices: await getAvailableModelChoices(true, config.apiKey),
+      initial: 0,
+    },
+    questionHandlers,
+  );
+  config.embeddingModel = embeddingModel;
+  config.dimensions = getDimensions(embeddingModel);
 
   return config;
 }

packages/create-llama/helpers/python.ts

@@ -0,0 +1,531 @@
+import fs from "fs/promises";
+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 {
+  InstallTemplateArgs,
+  ModelConfig,
+  TemplateDataSource,
+  TemplateVectorDB,
+} from "./types";
+
+interface Dependency {
+  name: string;
+  version?: string;
+  extras?: string[];
+  constraints?: Record<string, string>;
+}
+
+const getAdditionalDependencies = (
+  modelConfig: ModelConfig,
+  vectorDb?: TemplateVectorDB,
+  dataSources?: TemplateDataSource[],
+) => {
+  const dependencies: Dependency[] = [];
+
+  // Add vector db dependencies
+  switch (vectorDb) {
+    case "mongo": {
+      dependencies.push({
+        name: "llama-index-vector-stores-mongodb",
+        version: ">=0.3.2,<0.4.0",
+      });
+      break;
+    }
+    case "pg": {
+      dependencies.push({
+        name: "llama-index-vector-stores-postgres",
+        version: ">=0.3.2,<0.4.0",
+      });
+      break;
+    }
+    case "pinecone": {
+      dependencies.push({
+        name: "llama-index-vector-stores-pinecone",
+        version: ">=0.4.1,<0.5.0",
+        constraints: {
+          python: ">=3.11,<3.13",
+        },
+      });
+      break;
+    }
+    case "milvus": {
+      dependencies.push({
+        name: "llama-index-vector-stores-milvus",
+        version: ">=0.3.0,<0.4.0",
+      });
+      dependencies.push({
+        name: "pymilvus",
+        version: ">=2.4.4,<3.0.0",
+      });
+      break;
+    }
+    case "astra": {
+      dependencies.push({
+        name: "llama-index-vector-stores-astra-db",
+        version: ">=0.4.0,<0.5.0",
+      });
+      break;
+    }
+    case "qdrant": {
+      dependencies.push({
+        name: "llama-index-vector-stores-qdrant",
+        version: ">=0.4.0,<0.5.0",
+        constraints: {
+          python: ">=3.11,<3.13",
+        },
+      });
+      break;
+    }
+    case "chroma": {
+      dependencies.push({
+        name: "llama-index-vector-stores-chroma",
+        version: ">=0.4.0,<0.5.0",
+      });
+      dependencies.push({
+        name: "onnxruntime",
+        version: "<1.22.0",
+      });
+      break;
+    }
+    case "weaviate": {
+      dependencies.push({
+        name: "llama-index-vector-stores-weaviate",
+        version: ">=1.2.3,<2.0.0",
+      });
+      break;
+    }
+    case "llamacloud":
+      dependencies.push({
+        name: "llama-index-indices-managed-llama-cloud",
+        version: ">=0.6.3,<0.7.0",
+      });
+      break;
+  }
+
+  // Add data source dependencies
+  if (dataSources) {
+    for (const ds of dataSources) {
+      const dsType = ds?.type;
+      switch (dsType) {
+        case "file":
+          dependencies.push({
+            name: "docx2txt",
+            version: ">=0.8,<0.9",
+          });
+          break;
+        case "web":
+          dependencies.push({
+            name: "llama-index-readers-web",
+            version: ">=0.3.0,<0.4.0",
+          });
+          break;
+        case "db":
+          dependencies.push({
+            name: "llama-index-readers-database",
+            version: ">=0.3.0,<0.4.0",
+          });
+          dependencies.push({
+            name: "pymysql",
+            version: ">=1.1.0,<2.0.0",
+            extras: ["rsa"],
+          });
+          dependencies.push({
+            name: "psycopg2-binary",
+            version: ">=2.9.9,<3.0.0",
+          });
+          break;
+      }
+    }
+  }
+
+  switch (modelConfig.provider) {
+    case "ollama":
+      dependencies.push({
+        name: "llama-index-llms-ollama",
+        version: ">=0.5.0,<0.6.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-ollama",
+        version: ">=0.6.0,<0.7.0",
+      });
+      break;
+    case "openai":
+      dependencies.push({
+        name: "llama-index-llms-openai",
+        version: ">=0.3.2,<0.4.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-openai",
+        version: ">=0.3.1,<0.4.0",
+      });
+      break;
+    case "groq":
+      dependencies.push({
+        name: "llama-index-llms-groq",
+        version: ">=0.3.0,<0.4.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-fastembed",
+        version: ">=0.3.0,<0.4.0",
+      });
+      break;
+    case "anthropic":
+      dependencies.push({
+        name: "llama-index-llms-anthropic",
+        version: ">=0.6.0,<0.7.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-fastembed",
+        version: ">=0.3.0,<0.4.0",
+      });
+      break;
+    case "gemini":
+      dependencies.push({
+        name: "llama-index-llms-google-genai",
+        version: ">=0.2.0,<0.3.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-google-genai",
+        version: ">=0.2.0,<0.3.0",
+      });
+      break;
+    case "mistral":
+      dependencies.push({
+        name: "llama-index-llms-mistralai",
+        version: ">=0.4.0,<0.5.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-mistralai",
+        version: ">=0.3.0,<0.4.0",
+      });
+      break;
+    case "azure-openai":
+      dependencies.push({
+        name: "llama-index-llms-azure-openai",
+        version: ">=0.3.0,<0.4.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-azure-openai",
+        version: ">=0.3.0,<0.4.0",
+      });
+      break;
+    case "huggingface":
+      dependencies.push({
+        name: "llama-index-llms-huggingface",
+        version: ">=0.5.0,<0.6.0",
+      });
+      dependencies.push({
+        name: "llama-index-embeddings-huggingface",
+        version: ">=0.5.0,<0.6.0",
+      });
+      dependencies.push({
+        name: "optimum",
+        version: ">=1.23.3,<2.0.0",
+        extras: ["onnxruntime"],
+      });
+      break;
+    case "t-systems":
+      dependencies.push({
+        name: "llama-index-agent-openai",
+        version: ">=0.4.0,<0.5.0",
+      });
+      dependencies.push({
+        name: "llama-index-llms-openai-like",
+        version: ">=0.3.0,<0.4.0",
+      });
+      break;
+  }
+
+  // If app template is llama-index-server and CI and SERVER_PACKAGE_PATH is set,
+  // add @llamaindex/server to dependencies
+  if (process.env.SERVER_PACKAGE_PATH) {
+    dependencies.push({
+      name: "llama-index-server",
+      version: `@file://${process.env.SERVER_PACKAGE_PATH}`,
+    });
+  }
+
+  return dependencies;
+};
+
+export const addDependencies = async (
+  projectDir: string,
+  dependencies: Dependency[],
+) => {
+  if (dependencies.length === 0) return;
+
+  const FILENAME = "pyproject.toml";
+  try {
+    // 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.`,
+      );
+    }
+
+    // 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;
+        }
+      }
+    }
+
+    // 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
+    }
+  } catch (error) {
+    console.log(
+      `Error while updating dependencies for Poetry project file ${FILENAME}\n`,
+      error,
+    );
+  }
+};
+
+export const installPythonDependencies = () => {
+  if (isUvAvailable()) {
+    console.log(
+      `Installing Python dependencies using uv. This may take a while...`,
+    );
+    const installSuccessful = tryUvSync();
+    if (!installSuccessful) {
+      console.error(
+        red(
+          "Installing dependencies using uv failed. Please check the error log above and ensure uv is installed correctly.",
+        ),
+      );
+      process.exit(1);
+    }
+  } 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.`,
+      ),
+    );
+    process.exit(1);
+  }
+};
+
+const installLlamaIndexServerTemplate = async ({
+  root,
+  useCase,
+  useLlamaParse,
+  modelConfig,
+}: Pick<
+  InstallTemplateArgs,
+  "root" | "useCase" | "useLlamaParse" | "modelConfig"
+>) => {
+  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", "use-cases", "python", useCase),
+  });
+
+  // copy model provider settings to app folder
+  await copy("**", path.join(root, "app"), {
+    cwd: path.join(
+      templatesDir,
+      "components",
+      "providers",
+      "python",
+      modelConfig.provider,
+    ),
+  });
+
+  // Copy custom UI component code
+  await copy(`*`, path.join(root, "components"), {
+    parents: true,
+    cwd: path.join(templatesDir, "components", "ui", "use-cases", useCase),
+  });
+
+  // Copy layout components to layout folder in root
+  await copy("*", path.join(root, "layout"), {
+    parents: true,
+    cwd: path.join(templatesDir, "components", "ui", "layout"),
+  });
+
+  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", "use-cases", "python", useCase),
+    rename: assetRelocator,
+  });
+};
+
+export const installPythonTemplate = async ({
+  appName,
+  root,
+  template,
+  framework,
+  vectorDb,
+  postInstallAction,
+  modelConfig,
+  dataSources,
+  useLlamaParse,
+  useCase,
+}: Pick<
+  InstallTemplateArgs,
+  | "appName"
+  | "root"
+  | "template"
+  | "framework"
+  | "vectorDb"
+  | "postInstallAction"
+  | "modelConfig"
+  | "dataSources"
+  | "useLlamaParse"
+  | "useCase"
+>) => {
+  console.log("\nInitializing Python project with template:", template, "\n");
+  const templatePath = path.join(templatesDir, "types", template, framework);
+  await copy("**", root, {
+    parents: true,
+    cwd: templatePath,
+    rename: assetRelocator,
+  });
+
+  if (template === "llamaindexserver") {
+    await installLlamaIndexServerTemplate({
+      root,
+      useCase,
+      useLlamaParse,
+      modelConfig,
+    });
+  } else {
+    throw new Error(`Template ${template} not supported`);
+  }
+
+  console.log("Adding additional dependencies");
+  const addOnDependencies = getAdditionalDependencies(
+    modelConfig,
+    vectorDb,
+    dataSources,
+  );
+
+  await addDependencies(root, addOnDependencies);
+
+  if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
+    installPythonDependencies();
+  }
+};

packages/create-llama/helpers/run-app.ts

@@ -26,22 +26,13 @@ const createProcess = (
   });
 };
 
-export function runReflexApp(appPath: string, port: number) {
-  const commandArgs = [
-    "run",
-    "reflex",
-    "run",
-    "--frontend-port",
-    port.toString(),
-  ];
-  return createProcess("poetry", commandArgs, {
-    stdio: "inherit",
-    cwd: appPath,
-  });
-}
-
-export function runFastAPIApp(appPath: string, port: number) {
-  return createProcess("poetry", ["run", "dev"], {
+export function runFastAPIApp(
+  appPath: string,
+  port: number,
+  template: TemplateType,
+) {
+  const commandArgs = ["run", "fastapi", "dev", "--port", `${port}`];
+  return createProcess("uv", commandArgs, {
     stdio: "inherit",
     cwd: appPath,
     env: { ...process.env, APP_PORT: `${port}` },
@@ -64,16 +55,10 @@ export async function runApp(
 ): Promise<void> {
   try {
     // Start the app
-    const defaultPort =
-      framework === "nextjs" || template === "reflex" ? 3000 : 8000;
+    const defaultPort = framework === "nextjs" ? 3000 : 8000;
 
-    const appRunner =
-      template === "reflex"
-        ? runReflexApp
-        : framework === "fastapi"
-          ? runFastAPIApp
-          : runTSApp;
-    await appRunner(appPath, port || defaultPort);
+    const appRunner = framework === "fastapi" ? runFastAPIApp : runTSApp;
+    await appRunner(appPath, port || defaultPort, template);
   } catch (error) {
     console.error("Failed to run app:", error);
     throw error;

packages/create-llama/helpers/types.ts

@@ -1,5 +1,4 @@
 import { PackageManager } from "../helpers/get-pkg-manager";
-import { Tool } from "./tools";
 
 export type ModelProvider =
   | "openai"
@@ -19,15 +18,8 @@ export type ModelConfig = {
   dimensions: number;
   isConfigured(): boolean;
 };
-export type TemplateType =
-  | "streaming"
-  | "community"
-  | "llamapack"
-  | "multiagent"
-  | "reflex"
-  | "llamaindexserver";
+export type TemplateType = "llamaindexserver";
 export type TemplateFramework = "nextjs" | "express" | "fastapi";
-export type TemplateUI = "html" | "shadcn";
 export type TemplateVectorDB =
   | "none"
   | "mongo"
@@ -49,15 +41,22 @@ export type TemplateDataSource = {
   config: TemplateDataSourceConfig;
 };
 export type TemplateDataSourceType = "file" | "web" | "db";
-export type TemplateObservability = "none" | "traceloop" | "llamatrace";
 export type TemplateUseCase =
   | "financial_report"
-  | "blog"
   | "deep_research"
-  | "form_filling"
-  | "extractor"
-  | "contract_review"
-  | "agentic_rag";
+  | "agentic_rag"
+  | "code_generator"
+  | "document_generator"
+  | "hitl";
+
+export const ALL_USE_CASES: TemplateUseCase[] = [
+  "agentic_rag",
+  "deep_research",
+  "financial_report",
+  "code_generator",
+  "document_generator",
+  "hitl",
+];
 // Config for both file and folder
 export type FileSourceConfig =
   | {
@@ -83,31 +82,18 @@ export type TemplateDataSourceConfig =
   | WebSourceConfig
   | DbSourceConfig;
 
-export type CommunityProjectConfig = {
-  owner: string;
-  repo: string;
-  branch: string;
-  filePath?: string;
-};
-
 export interface InstallTemplateArgs {
   appName: string;
   root: string;
   packageManager: PackageManager;
-  isOnline: boolean;
   template: TemplateType;
   framework: TemplateFramework;
-  ui: TemplateUI;
   dataSources: TemplateDataSource[];
   modelConfig: ModelConfig;
   llamaCloudKey?: string;
-  useLlamaParse?: boolean;
-  communityProjectConfig?: CommunityProjectConfig;
-  llamapack?: string;
-  vectorDb?: TemplateVectorDB;
+  useLlamaParse: boolean;
+  vectorDb: TemplateVectorDB;
   port?: number;
-  postInstallAction?: TemplatePostInstallAction;
-  tools?: Tool[];
-  observability?: TemplateObservability;
-  useCase?: TemplateUseCase;
+  postInstallAction: TemplatePostInstallAction;
+  useCase: TemplateUseCase;
 }

packages/create-llama/helpers/typescript.ts

@@ -0,0 +1,286 @@
+import fs from "fs/promises";
+import os from "os";
+import path from "path";
+import { bold, cyan, red } from "picocolors";
+import { assetRelocator, copy } from "../helpers/copy";
+import { callPackageManager } from "../helpers/install";
+import { templatesDir } from "./dir";
+import { PackageManager } from "./get-pkg-manager";
+import { InstallTemplateArgs, ModelProvider, TemplateVectorDB } from "./types";
+
+const installLlamaIndexServerTemplate = async ({
+  root,
+  useCase,
+  vectorDb,
+  modelConfig,
+  dataSources,
+}: Pick<
+  InstallTemplateArgs,
+  "root" | "useCase" | "vectorDb" | "modelConfig" | "dataSources"
+>) => {
+  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);
+  }
+
+  // copy model provider settings to app folder
+  await copy("**", path.join(root, "src", "app"), {
+    cwd: path.join(
+      templatesDir,
+      "components",
+      "providers",
+      "typescript",
+      modelConfig.provider,
+    ),
+  });
+
+  await copy("**", path.join(root), {
+    cwd: path.join(
+      templatesDir,
+      "components",
+      "use-cases",
+      "typescript",
+      useCase,
+    ),
+    rename: assetRelocator,
+  });
+
+  // copy workflow UI components to components folder in root
+  await copy("*", path.join(root, "components"), {
+    parents: true,
+    cwd: path.join(templatesDir, "components", "ui", "use-cases", useCase),
+  });
+
+  // copy layout components to layout folder in root
+  await copy("*", path.join(root, "layout"), {
+    parents: true,
+    cwd: path.join(templatesDir, "components", "ui", "layout"),
+  });
+
+  // Override generate.ts if workflow use case doesn't use custom UI
+  if (vectorDb === "llamacloud") {
+    await copy("**", path.join(root, "src"), {
+      parents: true,
+      cwd: path.join(
+        templatesDir,
+        "components",
+        "vectordbs",
+        "llamaindexserver",
+        "llamacloud",
+        "typescript",
+      ),
+    });
+  }
+
+  // Simplify use case code
+  if (vectorDb === "none" && dataSources.length === 0) {
+    // use case without data sources doesn't use index.
+    // We don't need data.ts, generate.ts
+    await fs.rm(path.join(root, "src", "app", "data.ts"));
+    // TODO: split generate.ts into generate for index and generate for ui and remove generate for index here too
+    // then we can also remove it for llamacloud
+  }
+};
+
+/**
+ * Install a LlamaIndex internal template to a given `root` directory.
+ */
+export const installTSTemplate = async ({
+  appName,
+  root,
+  packageManager,
+  template,
+  framework,
+  vectorDb,
+  postInstallAction,
+  dataSources,
+  useCase,
+  modelConfig,
+}: InstallTemplateArgs) => {
+  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,
+  });
+
+  if (template === "llamaindexserver") {
+    await installLlamaIndexServerTemplate({
+      root,
+      useCase,
+      vectorDb,
+      modelConfig,
+      dataSources,
+    });
+  } else {
+    throw new Error(`Template ${template} not supported`);
+  }
+
+  const packageJson = await updatePackageJson({
+    root,
+    appName,
+    vectorDb,
+    modelConfig,
+  });
+
+  if (postInstallAction === "runApp" || postInstallAction === "dependencies") {
+    await installTSDependencies(packageJson, packageManager, true);
+  }
+};
+
+const providerDependencies: {
+  [key in ModelProvider]?: Record<string, string>;
+} = {
+  openai: {
+    "@llamaindex/openai": "~0.4.0",
+  },
+  gemini: {
+    "@llamaindex/google": "^0.2.0",
+  },
+  ollama: {
+    "@llamaindex/ollama": "^0.1.0",
+  },
+  mistral: {
+    "@llamaindex/mistral": "^0.2.0",
+  },
+  "azure-openai": {
+    "@llamaindex/openai": "^0.2.0",
+  },
+  groq: {
+    "@llamaindex/groq": "^0.0.61",
+    "@llamaindex/huggingface": "^0.1.0", // 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
+  },
+};
+
+const vectorDbDependencies: Record<TemplateVectorDB, Record<string, string>> = {
+  astra: {
+    "@llamaindex/astra": "^0.0.5",
+  },
+  chroma: {
+    "@llamaindex/chroma": "^0.0.5",
+  },
+  llamacloud: {},
+  milvus: {
+    "@zilliz/milvus2-sdk-node": "^2.4.6",
+    "@llamaindex/milvus": "^0.1.0",
+  },
+  mongo: {
+    mongodb: "6.7.0",
+    "@llamaindex/mongodb": "^0.0.5",
+  },
+  none: {},
+  pg: {
+    pg: "^8.12.0",
+    pgvector: "^0.2.0",
+    "@llamaindex/postgres": "^0.0.33",
+  },
+  pinecone: {
+    "@llamaindex/pinecone": "^0.0.5",
+  },
+  qdrant: {
+    "@qdrant/js-client-rest": "^1.11.0",
+    "@llamaindex/qdrant": "^0.1.0",
+  },
+  weaviate: {
+    "@llamaindex/weaviate": "^0.0.5",
+  },
+};
+
+async function updatePackageJson({
+  root,
+  appName,
+  vectorDb,
+  modelConfig,
+}: Pick<
+  InstallTemplateArgs,
+  "root" | "appName" | "vectorDb" | "modelConfig"
+>): Promise<any> {
+  const packageJsonFile = path.join(root, "package.json");
+  const packageJson: any = JSON.parse(
+    await fs.readFile(packageJsonFile, "utf8"),
+  );
+  packageJson.name = appName;
+  packageJson.version = "0.1.0";
+
+  packageJson.dependencies = {
+    ...packageJson.dependencies,
+    "@llamaindex/readers": "~3.1.4",
+  };
+
+  if (vectorDb && vectorDb in vectorDbDependencies) {
+    packageJson.dependencies = {
+      ...packageJson.dependencies,
+      ...vectorDbDependencies[vectorDb],
+    };
+  }
+
+  if (modelConfig.provider && modelConfig.provider in providerDependencies) {
+    packageJson.dependencies = {
+      ...packageJson.dependencies,
+      ...providerDependencies[modelConfig.provider],
+    };
+  }
+
+  // if having custom server package tgz file, use it for testing @llamaindex/server
+  const serverPackagePath = process.env.SERVER_PACKAGE_PATH;
+  if (serverPackagePath) {
+    const relativePath = path.relative(process.cwd(), serverPackagePath);
+    packageJson.dependencies = {
+      ...packageJson.dependencies,
+      "@llamaindex/server": `file:${relativePath}`,
+    };
+  }
+
+  await fs.writeFile(
+    packageJsonFile,
+    JSON.stringify(packageJson, null, 2) + os.EOL,
+  );
+
+  return packageJson;
+}
+
+async function installTSDependencies(
+  packageJson: any,
+  packageManager: PackageManager,
+  isOnline: boolean,
+): Promise<void> {
+  console.log("\nInstalling dependencies:");
+  for (const dependency in packageJson.dependencies)
+    console.log(`- ${cyan(dependency)}`);
+
+  console.log("\nInstalling devDependencies:");
+  for (const dependency in packageJson.devDependencies)
+    console.log(`- ${cyan(dependency)}`);
+
+  console.log();
+
+  await callPackageManager(packageManager, isOnline).catch((error) => {
+    console.error("Failed to install TS dependencies. Exiting...");
+    process.exit(1);
+  });
+}

packages/create-llama/helpers/uv.ts

@@ -0,0 +1,42 @@
+// 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/helpers/validate-pkg.ts

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

packages/create-llama/index.ts

@@ -1,4 +1,3 @@
-/* eslint-disable import/no-extraneous-dependencies */
 import { execSync } from "child_process";
 import { Command } from "commander";
 import fs from "fs";
@@ -8,12 +7,10 @@ import prompts from "prompts";
 import terminalLink from "terminal-link";
 import checkForUpdate from "update-check";
 import { createApp } from "./create-app";
-import { EXAMPLE_FILE, getDataSources } from "./helpers/datasources";
 import { getPkgManager } from "./helpers/get-pkg-manager";
 import { isFolderEmpty } from "./helpers/is-folder-empty";
 import { initializeGlobalAgent } from "./helpers/proxy";
 import { runApp } from "./helpers/run-app";
-import { getTools } from "./helpers/tools";
 import { validateNpmName } from "./helpers/validate-pkg";
 import packageJson from "./package.json";
 import { askQuestions } from "./questions/index";
@@ -57,13 +54,6 @@ const program = new Command(packageJson.name)
     `
 
   Explicitly tell the CLI to bootstrap the application using Yarn
-`,
-  )
-  .option(
-    "--template <template>",
-    `
-
-  Select a template to bootstrap the application with.
 `,
   )
   .option(
@@ -71,62 +61,6 @@ const program = new Command(packageJson.name)
     `
 
   Select a framework to bootstrap the application with.
-`,
-  )
-  .option(
-    "--files <path>",
-    `
-  
-  Specify the path to a local file or folder for chatting.
-`,
-  )
-  .option(
-    "--example-file",
-    `
-
-  Select to use an example PDF as data source.
-`,
-  )
-  .option(
-    "--web-source <url>",
-    `
-  
-  Specify a website URL to use as a data source.
-`,
-  )
-  .option(
-    "--db-source <connection-string>",
-    `
-  
-  Specify a database connection string to use as a data source.
-`,
-  )
-  .option(
-    "--open-ai-key <key>",
-    `
-
-  Provide an OpenAI API key.
-`,
-  )
-  .option(
-    "--ui <ui>",
-    `
-
-  Select a UI to bootstrap the application with.
-`,
-  )
-  .option(
-    "--frontend",
-    `
-
-  Generate a frontend for your backend.
-`,
-  )
-  .option(
-    "--no-frontend",
-    `
-
-  Do not generate a frontend for your backend.
 `,
   )
   .option(
@@ -148,27 +82,6 @@ const program = new Command(packageJson.name)
     `
 
   Select which vector database you would like to use, such as 'none', 'pg' or 'mongo'. The default option is not to use a vector database and use the local filesystem instead ('none').
-`,
-  )
-  .option(
-    "--tools <tools>",
-    `
-
-  Specify the tools you want to use by providing a comma-separated list. For example, 'wikipedia.WikipediaToolSpec,google.GoogleSearchToolSpec'. Use 'none' to not using any tools.
-`,
-    (tools, _) => {
-      if (tools === "none") {
-        return [];
-      } else {
-        return getTools(tools.split(","));
-      }
-    },
-  )
-  .option(
-    "--use-llama-parse",
-    `
-
-  Enable LlamaParse.
 `,
   )
   .option(
@@ -178,26 +91,12 @@ const program = new Command(packageJson.name)
   Provide a LlamaCloud API key.
 `,
   )
-  .option(
-    "--observability <observability>",
-    `
-    
-  Specify observability tools to use. Eg: none, opentelemetry
-`,
-  )
+
   .option(
     "--ask-models",
     `
 
   Allow interactive selection of LLM and embedding models of different model providers.
-`,
-    false,
-  )
-  .option(
-    "--pro",
-    `
-
-  Allow interactive selection of all features.
 `,
     false,
   )
@@ -205,7 +104,7 @@ const program = new Command(packageJson.name)
     "--use-case <useCase>",
     `
 
-  Select which use case to use for the multi-agent template (e.g: financial_report, blog).
+  Select which use case to use for the template (e.g: financial_report, blog).
 `,
   )
   .allowUnknownOption()
@@ -213,42 +112,6 @@ const program = new Command(packageJson.name)
 
 const options = program.opts();
 
-if (
-  process.argv.includes("--no-llama-parse") ||
-  options.template === "reflex"
-) {
-  options.useLlamaParse = false;
-}
-if (process.argv.includes("--no-files")) {
-  options.dataSources = [];
-} else if (process.argv.includes("--example-file")) {
-  options.dataSources = getDataSources(options.files, options.exampleFile);
-} else if (process.argv.includes("--llamacloud")) {
-  options.dataSources = [EXAMPLE_FILE];
-  options.vectorDb = "llamacloud";
-} else if (process.argv.includes("--web-source")) {
-  options.dataSources = [
-    {
-      type: "web",
-      config: {
-        baseUrl: options.webSource,
-        prefix: options.webSource,
-        depth: 1,
-      },
-    },
-  ];
-} else if (process.argv.includes("--db-source")) {
-  options.dataSources = [
-    {
-      type: "db",
-      config: {
-        uri: options.dbSource,
-        queries: options.dbQuery || "SELECT * FROM mytable",
-      },
-    },
-  ];
-}
-
 const packageManager = !!options.useNpm
   ? "npm"
   : !!options.usePnpm
@@ -257,6 +120,9 @@ const packageManager = !!options.useNpm
       ? "yarn"
       : getPkgManager();
 
+// options above must use all the properties of QuestionArgs
+const cliArgs = options as unknown as QuestionArgs;
+
 async function run(): Promise<void> {
   if (typeof projectPath === "string") {
     projectPath = projectPath.trim();
@@ -320,7 +186,7 @@ async function run(): Promise<void> {
     process.exit(1);
   }
 
-  const answers = await askQuestions(options as unknown as QuestionArgs);
+  const answers = await askQuestions(cliArgs);
 
   await createApp({
     ...answers,

packages/create-llama/package.json

@@ -0,0 +1,75 @@
+{
+  "name": "create-llama",
+  "version": "0.6.1",
+  "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:ts": "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",
+    "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/playwright.config.ts

@@ -1,4 +1,3 @@
-/* eslint-disable turbo/no-undeclared-env-vars */
 import { defineConfig, devices } from "@playwright/test";
 
 export default defineConfig({

packages/create-llama/questions/index.ts

@@ -0,0 +1,149 @@
+import prompts from "prompts";
+import { askModelConfig } from "../helpers/providers";
+import {
+  TemplateFramework,
+  TemplateUseCase,
+  TemplateVectorDB,
+} from "../helpers/types";
+import { QuestionArgs, QuestionResults } from "./types";
+import { useCaseConfiguration } from "./usecases";
+import { askPostInstallAction, questionHandlers } from "./utils";
+
+export const askQuestions = async (
+  args: QuestionArgs,
+): Promise<QuestionResults> => {
+  const {
+    useCase: useCaseFromArgs,
+    framework: frameworkFromArgs,
+    llamaCloudKey: llamaCloudKeyFromArgs,
+    vectorDb: vectorDbFromArgs,
+    postInstallAction: postInstallActionFromArgs,
+    askModels: askModelsFromArgs,
+  } = args;
+
+  const { useCase, framework } = await prompts(
+    [
+      {
+        type: useCaseFromArgs ? null : "select",
+        name: "useCase",
+        message: "What use case do you want to build?",
+        choices: [
+          {
+            title: "Agentic RAG",
+            value: "agentic_rag",
+            description:
+              "Chatbot that answers questions based on provided documents.",
+          },
+          {
+            title: "Financial Report",
+            value: "financial_report",
+            description:
+              "Agent that analyzes data and generates visualizations by using a code interpreter.",
+          },
+          {
+            title: "Deep Research",
+            value: "deep_research",
+            description:
+              "Researches and analyzes provided documents from multiple perspectives, generating a comprehensive report with citations to support key findings and insights.",
+          },
+          {
+            title: "Code Generator",
+            value: "code_generator",
+            description: "Build a Vercel v0 styled code generator.",
+          },
+          {
+            title: "Document Generator",
+            value: "document_generator",
+            description: "Build a OpenAI canvas-styled document generator.",
+          },
+          {
+            title: "Human in the Loop",
+            value: "hitl",
+            description:
+              "Build a CLI command workflow that is reviewed by a human before execution",
+          },
+        ],
+        initial: 0,
+      },
+      {
+        type: frameworkFromArgs ? null : "select",
+        name: "framework",
+        message: "What language do you want to use?",
+        choices: [
+          { title: "Python (FastAPI)", value: "fastapi" },
+          { title: "Typescript (NextJS)", value: "nextjs" },
+        ],
+        initial: 0,
+      },
+    ],
+    questionHandlers,
+  );
+
+  const finalUseCase = (useCaseFromArgs ?? useCase) as TemplateUseCase;
+  const finalFramework = (frameworkFromArgs ?? framework) as TemplateFramework;
+  if (!finalUseCase) {
+    throw new Error("Use case is required");
+  }
+  if (!finalFramework) {
+    throw new Error("Framework is required");
+  }
+
+  // lookup configuration for the use case
+  const useCaseConfig = useCaseConfiguration[finalUseCase];
+
+  // Ask for model provider
+  let modelConfig = useCaseConfig.modelConfig;
+  if (askModelsFromArgs) {
+    modelConfig = await askModelConfig({
+      framework: finalFramework,
+    });
+  }
+
+  // Ask for LlamaCloud
+  let llamaCloudKey = llamaCloudKeyFromArgs ?? process.env.LLAMA_CLOUD_API_KEY;
+  let vectorDb: TemplateVectorDB = vectorDbFromArgs ?? "none";
+  if (!vectorDbFromArgs && useCaseConfig.dataSources) {
+    const { useLlamaCloud } = await prompts(
+      {
+        type: "toggle",
+        name: "useLlamaCloud",
+        message: "Do you want to use LlamaCloud?",
+        active: "Yes",
+        inactive: "No",
+        initial: false,
+      },
+      questionHandlers,
+    );
+    if (useLlamaCloud && !llamaCloudKey) {
+      const { llamaCloudKey: llamaCloudKeyFromPrompt } = await prompts(
+        {
+          type: "text",
+          name: "llamaCloudKey",
+          message:
+            "Please provide your LlamaCloud API key (leave blank to skip):",
+        },
+        questionHandlers,
+      );
+      llamaCloudKey = llamaCloudKeyFromPrompt;
+    }
+    vectorDb = useLlamaCloud ? "llamacloud" : "none";
+  }
+
+  const result = {
+    ...useCaseConfig,
+    framework: finalFramework,
+    useCase: finalUseCase,
+    modelConfig,
+    llamaCloudKey,
+    useLlamaParse: vectorDb === "llamacloud",
+    vectorDb,
+  };
+
+  const postInstallAction =
+    postInstallActionFromArgs ?? (await askPostInstallAction(result));
+
+  return {
+    ...result,
+    postInstallAction,
+  };
+};

packages/create-llama/questions/types.ts

@@ -0,0 +1,22 @@
+import { InstallAppArgs } from "../create-app";
+import {
+  TemplateFramework,
+  TemplatePostInstallAction,
+  TemplateUseCase,
+  TemplateVectorDB,
+} from "../helpers";
+
+export type QuestionResults = Omit<
+  InstallAppArgs,
+  "appPath" | "packageManager"
+>;
+
+export type QuestionArgs = {
+  useCase?: TemplateUseCase;
+  framework?: TemplateFramework;
+  askModels?: boolean;
+  llamaCloudKey?: string;
+  port?: number;
+  postInstallAction?: TemplatePostInstallAction;
+  vectorDb?: TemplateVectorDB;
+};

packages/create-llama/questions/usecases.ts

@@ -0,0 +1,42 @@
+import { EXAMPLE_10K_SEC_FILES, EXAMPLE_FILE } from "../helpers/datasources";
+import { getGpt41ModelConfig } from "../helpers/models";
+import { ModelConfig, TemplateUseCase } from "../helpers/types";
+import { QuestionResults } from "./types";
+
+export const useCaseConfiguration: Record<
+  TemplateUseCase,
+  Pick<QuestionResults, "template" | "dataSources"> & {
+    modelConfig: ModelConfig;
+  }
+> = {
+  agentic_rag: {
+    template: "llamaindexserver",
+    dataSources: [EXAMPLE_FILE],
+    modelConfig: getGpt41ModelConfig(),
+  },
+  financial_report: {
+    template: "llamaindexserver",
+    dataSources: EXAMPLE_10K_SEC_FILES,
+    modelConfig: getGpt41ModelConfig(),
+  },
+  deep_research: {
+    template: "llamaindexserver",
+    dataSources: EXAMPLE_10K_SEC_FILES,
+    modelConfig: getGpt41ModelConfig(),
+  },
+  code_generator: {
+    template: "llamaindexserver",
+    dataSources: [],
+    modelConfig: getGpt41ModelConfig(),
+  },
+  document_generator: {
+    template: "llamaindexserver",
+    dataSources: [],
+    modelConfig: getGpt41ModelConfig(),
+  },
+  hitl: {
+    template: "llamaindexserver",
+    dataSources: [],
+    modelConfig: getGpt41ModelConfig(),
+  },
+};

packages/create-llama/questions/utils.ts

@@ -4,7 +4,6 @@ import path from "path";
 import { red } from "picocolors";
 import prompts from "prompts";
 import { TemplateDataSourceType, TemplatePostInstallAction } from "../helpers";
-import { toolsRequireConfig } from "../helpers/tools";
 import { QuestionResults } from "./types";
 
 export const supportedContextFileTypes = [
@@ -127,7 +126,7 @@ export const questionHandlers = {
 
 // Ask for next action after installation
 export async function askPostInstallAction(
-  args: QuestionResults,
+  args: Omit<QuestionResults, "postInstallAction">,
 ): Promise<TemplatePostInstallAction> {
   const actionChoices = [
     {
@@ -144,19 +143,14 @@ export async function askPostInstallAction(
     },
   ];
 
-  const modelConfigured = !args.llamapack && args.modelConfig.isConfigured();
+  const modelConfigured = args.modelConfig.isConfigured();
   // If using LlamaParse, require LlamaCloud API key
   const llamaCloudKeyConfigured = args.useLlamaParse
     ? args.llamaCloudKey || process.env["LLAMA_CLOUD_API_KEY"]
     : true;
   const hasVectorDb = args.vectorDb && args.vectorDb !== "none";
   // Can run the app if all tools do not require configuration
-  if (
-    !hasVectorDb &&
-    modelConfigured &&
-    llamaCloudKeyConfigured &&
-    !toolsRequireConfig(args.tools)
-  ) {
+  if (!hasVectorDb && modelConfigured && llamaCloudKeyConfigured) {
     actionChoices.push({
       title: "Generate code, install dependencies, and run the app (~2 min)",
       value: "runApp",

packages/create-llama/templates/components/agents/python/blog/README-template.md

@@ -19,20 +19,20 @@ First, setup the environment with poetry:
 > **_Note:_** This step is not needed if you are using the dev-container.
 
 ```shell
-poetry install
+uv sync
 ```
 
 Then check the parameters that have been pre-configured in the `.env` file in this directory. (E.g. you might need to configure an `OPENAI_API_KEY` if you're using OpenAI as model provider).
 Second, generate the embeddings of the documents in the `./data` directory:
 
 ```shell
-poetry run generate
+uv run generate
 ```
 
 Third, run the development server:
 
 ```shell
-poetry run dev
+uv run dev
 ```
 
 Per default, the example is using the explicit workflow. You can change the example by setting the `EXAMPLE_TYPE` environment variable to `choreography` or `orchestrator`.
@@ -52,7 +52,7 @@ Open [http://localhost:8000](http://localhost:8000) with your browser to start t
 To start the app optimized for **production**, run:
 
 ```
-poetry run prod
+uv run prod
 ```
 
 ## Deployments