Full Code of bertrandamobi/AI-Workflow-Engine for AI

Repository: bertrandamobi/AI-Workflow-Engine
Branch: main
Commit: 478a9dd05496
Files: 27
Total size: 21.2 KB

Directory structure:
gitextract_415u4r91/

├── .github/
│   └── workflows/
│       └── deploy-pages.yml
├── .gitignore
├── .gitkeep
├── README.md
├── backend/
│   ├── Dockerfile
│   ├── app/
│   │   ├── __init__.py
│   │   ├── api/
│   │   │   └── workflows.py
│   │   ├── core/
│   │   │   ├── config.py
│   │   │   └── logging.py
│   │   ├── graph/
│   │   │   └── workflow.py
│   │   ├── main.py
│   │   ├── schemas/
│   │   │   └── workflow.py
│   │   └── services/
│   │       ├── llm_service.py
│   │       └── vector_store.py
│   ├── pyproject.toml
│   └── tests/
│       └── test_health.py
├── docker-compose.yml
└── frontend/
    ├── Dockerfile
    ├── next.config.mjs
    ├── package.json
    ├── src/
    │   ├── app/
    │   │   ├── layout.tsx
    │   │   ├── page.tsx
    │   │   └── styles.css
    │   ├── components/
    │   │   └── WorkflowForm.tsx
    │   ├── lib/
    │   │   └── api.ts
    │   └── types/
    │       └── workflow.ts
    └── tsconfig.json

================================================
FILE CONTENTS
================================================

================================================
FILE: .github/workflows/deploy-pages.yml
================================================
name: Deploy Next.js static site to Pages

on:
  push:
    branches: ["main"]
  workflow_dispatch:

permissions:
  contents: read
  pages: write
  id-token: write

concurrency:
  group: "pages"
  cancel-in-progress: true

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Setup Node
        uses: actions/setup-node@v4
        with:
          node-version: 22

      - name: Install dependencies
        run: npm install
        working-directory: frontend

      - name: Build static site
        run: npm run build
        working-directory: frontend

      - name: Upload Pages artifact
        uses: actions/upload-pages-artifact@v3
        with:
          path: frontend/out

  deploy:
    environment:
      name: github-pages
      url: ${{ steps.deployment.outputs.page_url }}
    runs-on: ubuntu-latest
    needs: build
    steps:
      - name: Deploy to GitHub Pages
        id: deployment
        uses: actions/deploy-pages@v4


================================================
FILE: .gitignore
================================================
# Python
__pycache__/
*.py[cod]
*.so
.venv/
.pytest_cache/
.mypy_cache/
.coverage
htmlcov/

# Node
node_modules/
.next/
dist/
coverage/

# Environment
.env
.env.local

# IDE
.vscode/
.idea/
.DS_Store

# Build
build/
*.log


================================================
FILE: .gitkeep
================================================


================================================
FILE: README.md
================================================
# Agentic AI Workflow Engine

A production-style, full-stack orchestration platform for autonomous market analysis workflows using **LangGraph**, **OpenAI**, and **vector retrieval patterns**.

## Architecture Overview

- **Backend**: FastAPI service exposing workflow execution APIs.
- **Orchestration**: LangGraph state-machine graph with explicit node transitions (`research -> analysis -> recommendations`).
- **LLM Layer**: OpenAI chat model abstraction with configurable model selection.
- **Retrieval Layer**: Chroma-based vector service abstraction for extensible RAG workflows.
- **Frontend**: Next.js dashboard to trigger workflows and inspect results.
- **Runtime**: Docker Compose for local full-stack bootstrapping.

## Repository Structure

```text
backend/
  app/
    api/           # HTTP routers
    core/          # config + logging
    graph/         # LangGraph workflow definitions
    schemas/       # API contracts
    services/      # LLM and vector abstractions
  tests/
frontend/
  src/
    app/           # Next.js pages
    components/    # UI components
    lib/           # API clients
    types/         # shared types
infra/             # reserved for IaC modules
scripts/           # tooling scripts
docs/              # architecture and ADRs
```

## Local Development

### 1) Prerequisites
- Python 3.11+
- Node 22+
- Docker + Docker Compose

### 2) Configure environment

```bash
cp .env.example .env
```

Populate at least:
- `OPENAI_API_KEY`
- `OPENAI_MODEL`

### 3) Run backend

```bash
cd backend
python -m venv .venv && source .venv/bin/activate
pip install -e .
uvicorn app.main:app --reload --port 8000
```

### 4) Run frontend

```bash
cd frontend
npm install
npm run dev
```

### 5) Run full stack with Docker

```bash
docker compose up --build
```

## API Contract

### POST `/workflows/market-analysis`

Request:
```json
{
  "company": "NVIDIA",
  "objectives": ["Competitive positioning", "Pricing strategy"],
  "context": {}
}
```

Response:
```json
{
  "workflow_id": "uuid",
  "status": "completed",
  "summary": "...",
  "artifacts": {
    "research_notes": "...",
    "recommendations": "..."
  }
}
```

## Engineering Characteristics

- Clear separation of concerns (API / orchestration / services / schema boundaries).
- Strong typing across Python and TypeScript.
- Environment-driven configuration with explicit defaults.
- Dockerized runtime and clean onboarding docs.
- Test scaffold for API health and easy extension for integration tests.

## Scalability Roadmap

1. Add task queue + worker runtime (Celery/Arq) for asynchronous workflow runs.
2. Persist workflow state transitions in SQLModel-backed store.
3. Add observability stack (OpenTelemetry traces, structured logs, metrics).
4. Add multi-tenant authn/authz and policy-enforced tool access.
5. Expand graph with conditional branches and evaluator nodes.

## Quality Gates (recommended)

Backend:
```bash
pytest
ruff check .
mypy app
```

Frontend:
```bash
npm run lint
npm run build
```

## GitHub Pages Demo Deployment

This repository includes a ready-to-use workflow at `.github/workflows/deploy-pages.yml`.

1. Push your code to `main`.
2. In GitHub: **Settings → Pages → Source = GitHub Actions**.
3. Run workflow (push or manual dispatch).

The build publishes `frontend/out/` to Pages.

### Why previous workflow failed

If `actions/setup-node` uses npm caching without a lockfile path that exists, it fails with:
`Some specified paths were not resolved, unable to cache dependencies.`

This repository workflow avoids that issue by not requiring `cache-dependency-path`.


================================================
FILE: backend/Dockerfile
================================================
FROM python:3.11-slim
WORKDIR /app
COPY pyproject.toml ./
RUN pip install --no-cache-dir -U pip && pip install --no-cache-dir .
COPY app ./app
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]


================================================
FILE: backend/app/__init__.py
================================================


================================================
FILE: backend/app/api/workflows.py
================================================
from fastapi import APIRouter

from app.graph.workflow import run_workflow
from app.schemas.workflow import WorkflowRequest, WorkflowResponse

router = APIRouter(prefix="/workflows", tags=["workflows"])


@router.post("/market-analysis", response_model=WorkflowResponse)
def execute_market_analysis(payload: WorkflowRequest) -> WorkflowResponse:
    result = run_workflow(payload.company, payload.objectives)
    return WorkflowResponse(**result)


================================================
FILE: backend/app/core/config.py
================================================
from functools import lru_cache

from pydantic_settings import BaseSettings, SettingsConfigDict


class Settings(BaseSettings):
    model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8", extra="ignore")

    app_env: str = "development"
    app_host: str = "0.0.0.0"
    app_port: int = 8000
    log_level: str = "INFO"

    openai_api_key: str = ""
    openai_model: str = "gpt-4.1"

    vector_store_provider: str = "chroma"
    chroma_persist_directory: str = "./data/chroma"

    database_url: str = "sqlite:///./workflow_engine.db"


@lru_cache(maxsize=1)
def get_settings() -> Settings:
    return Settings()


================================================
FILE: backend/app/core/logging.py
================================================
import logging


def configure_logging(level: str = "INFO") -> None:
    logging.basicConfig(
        level=getattr(logging, level.upper(), logging.INFO),
        format="%(asctime)s | %(levelname)s | %(name)s | %(message)s",
    )


================================================
FILE: backend/app/graph/workflow.py
================================================
from typing import TypedDict
from uuid import uuid4

from langgraph.graph import END, StateGraph

from app.services.llm_service import LLMService


class MarketState(TypedDict):
    company: str
    objectives: list[str]
    research_notes: str
    analysis: str
    recommendations: str


llm = LLMService()


def research_node(state: MarketState) -> MarketState:
    prompt = f"Research top market dynamics for {state['company']}."
    state["research_notes"] = llm.summarize(prompt)
    return state


def analysis_node(state: MarketState) -> MarketState:
    prompt = (
        f"Use notes to produce strategic analysis for {state['company']}: {state['research_notes']}"
    )
    state["analysis"] = llm.summarize(prompt)
    return state


def recommendations_node(state: MarketState) -> MarketState:
    prompt = (
        f"Provide 5 executive recommendations for {state['company']} based on {state['analysis']}"
    )
    state["recommendations"] = llm.summarize(prompt)
    return state


def build_market_workflow():
    graph = StateGraph(MarketState)
    graph.add_node("research", research_node)
    graph.add_node("analysis", analysis_node)
    graph.add_node("recommendations", recommendations_node)

    graph.set_entry_point("research")
    graph.add_edge("research", "analysis")
    graph.add_edge("analysis", "recommendations")
    graph.add_edge("recommendations", END)

    return graph.compile()


def run_workflow(company: str, objectives: list[str]) -> dict:
    workflow = build_market_workflow()
    workflow_id = str(uuid4())

    state: MarketState = {
        "company": company,
        "objectives": objectives,
        "research_notes": "",
        "analysis": "",
        "recommendations": "",
    }
    result = workflow.invoke(state)
    return {
        "workflow_id": workflow_id,
        "status": "completed",
        "summary": result["analysis"],
        "artifacts": {
            "research_notes": result["research_notes"],
            "recommendations": result["recommendations"],
        },
    }


================================================
FILE: backend/app/main.py
================================================
from fastapi import FastAPI

from app.api.workflows import router as workflow_router
from app.core.config import get_settings
from app.core.logging import configure_logging

settings = get_settings()
configure_logging(settings.log_level)

app = FastAPI(
    title="Agentic AI Workflow Engine",
    description="Enterprise orchestration engine powered by LangGraph",
    version="0.1.0",
)
app.include_router(workflow_router)


@app.get("/health")
def health_check() -> dict[str, str]:
    return {"status": "ok"}


================================================
FILE: backend/app/schemas/workflow.py
================================================
from typing import Any

from pydantic import BaseModel, Field


class WorkflowRequest(BaseModel):
    company: str = Field(..., description="Company to analyze")
    objectives: list[str] = Field(default_factory=list)
    context: dict[str, Any] = Field(default_factory=dict)


class WorkflowResponse(BaseModel):
    workflow_id: str
    status: str
    summary: str
    artifacts: dict[str, Any]


================================================
FILE: backend/app/services/llm_service.py
================================================
from langchain_openai import ChatOpenAI

from app.core.config import get_settings


class LLMService:
    def __init__(self) -> None:
        settings = get_settings()
        self.client = ChatOpenAI(model=settings.openai_model, api_key=settings.openai_api_key)

    def summarize(self, prompt: str) -> str:
        response = self.client.invoke(prompt)
        return response.content


================================================
FILE: backend/app/services/vector_store.py
================================================
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

from app.core.config import get_settings


class VectorStoreService:
    def __init__(self) -> None:
        settings = get_settings()
        self._embeddings = OpenAIEmbeddings(api_key=settings.openai_api_key)
        self._store = Chroma(
            persist_directory=settings.chroma_persist_directory,
            embedding_function=self._embeddings,
        )

    def similarity_search(self, query: str, k: int = 4):
        return self._store.similarity_search(query, k=k)


================================================
FILE: backend/pyproject.toml
================================================
[project]
name = "agentic-ai-workflow-engine"
version = "0.1.0"
description = "LangGraph-based AI orchestration system for autonomous market analysis"
readme = "README.md"
requires-python = ">=3.11"
dependencies = [
  "fastapi>=0.115.0",
  "uvicorn[standard]>=0.30.0",
  "pydantic-settings>=2.6.0",
  "langgraph>=0.2.0",
  "langchain>=0.3.0",
  "langchain-openai>=0.2.0",
  "langchain-community>=0.3.0",
  "chromadb>=0.5.0",
  "sqlmodel>=0.0.22",
  "httpx>=0.27.0",
]

[project.optional-dependencies]
dev = ["pytest>=8.3.0", "ruff>=0.7.0", "mypy>=1.12.0"]

[tool.ruff]
line-length = 100
target-version = "py311"

[tool.pytest.ini_options]
pythonpath = ["."]


================================================
FILE: backend/tests/test_health.py
================================================
from fastapi.testclient import TestClient

from app.main import app


def test_health_check() -> None:
    client = TestClient(app)
    response = client.get("/health")
    assert response.status_code == 200
    assert response.json() == {"status": "ok"}


================================================
FILE: docker-compose.yml
================================================
version: "3.9"
services:
  api:
    build:
      context: ./backend
    env_file:
      - .env
    ports:
      - "8000:8000"
    volumes:
      - ./backend:/app
      - ./data:/app/data
  web:
    build:
      context: ./frontend
    ports:
      - "3000:3000"
    environment:
      - NEXT_PUBLIC_API_BASE_URL=http://api:8000
    depends_on:
      - api


================================================
FILE: frontend/Dockerfile
================================================
FROM node:22-alpine
WORKDIR /app
COPY package.json ./
RUN npm install
COPY . .
CMD ["npm", "run", "dev", "--", "-H", "0.0.0.0", "-p", "3000"]


================================================
FILE: frontend/next.config.mjs
================================================
const isGithubActions = process.env.GITHUB_ACTIONS === "true";
const repoName = "AI-Workflow-Engine";

const nextConfig = {
  reactStrictMode: true,
  output: "export",
  images: {
    unoptimized: true,
  },
  basePath: isGithubActions ? `/${repoName}` : "",
  assetPrefix: isGithubActions ? `/${repoName}/` : "",
};

export default nextConfig;


================================================
FILE: frontend/package.json
================================================
{
  "name": "workflow-engine-web",
  "private": true,
  "version": "0.1.0",
  "scripts": {
    "dev": "next dev",
    "build": "next build",
    "start": "next start",
    "lint": "next lint"
  },
  "dependencies": {
    "next": "15.0.0",
    "react": "18.2.0",
    "react-dom": "18.2.0"
  },
  "devDependencies": {
    "typescript": "5.6.3",
    "@types/react": "18.3.12",
    "@types/node": "22.8.1",
    "eslint": "9.13.0",
    "eslint-config-next": "15.0.0"
  }
}


================================================
FILE: frontend/src/app/layout.tsx
================================================
import type { Metadata } from "next";

export const metadata: Metadata = {
  title: "Agentic AI Workflow Engine Demo",
  description: "Senior-level orchestration demo UI for LangGraph-powered workflows",
};

export default function RootLayout({ children }: { children: React.ReactNode }) {
  return (
    <html lang="en">
      <body>{children}</body>
    </html>
  );
}


================================================
FILE: frontend/src/app/page.tsx
================================================
import WorkflowForm from "../components/WorkflowForm";
import "./styles.css";

export default function HomePage() {
  return (
    <main className="page-shell">
      <div className="aurora aurora-a" />
      <div className="aurora aurora-b" />
      <div className="noise-overlay" />

      <header className="hero">
        <p className="eyebrow">LangGraph • OpenAI • Retrieval • Orchestration</p>
        <h1>Agentic AI Workflow Engine</h1>
        <p className="hero-copy">
          Autonomous market analysis with stateful graph execution, retriever-augmented context,
          and executive-grade recommendation synthesis.
        </p>
        <div className="hero-metrics">
          <article>
            <span>3</span>
            <p>Orchestration Nodes</p>
          </article>
          <article>
            <span>1 Click</span>
            <p>Run Strategic Analysis</p>
          </article>
          <article>
            <span>Typed</span>
            <p>End-to-End Contracts</p>
          </article>
        </div>
      </header>

      <section className="content-grid">
        <article className="panel elevated">
          <h2>Workflow Studio</h2>
          <p className="panel-copy">
            Trigger a complete research → analysis → recommendation cycle. Designed for portfolio
            strategy, product intelligence, and competitive planning.
          </p>
          <WorkflowForm />
        </article>

        <aside className="panel glass">
          <h2>System Highlights</h2>
          <ul>
            <li>Deterministic LangGraph state machine orchestration.</li>
            <li>OpenAI model abstraction for configurable inference strategy.</li>
            <li>Vector retrieval-ready service layer for contextual augmentation.</li>
            <li>Containerized full-stack runtime for local and cloud demos.</li>
          </ul>
          <div className="pulse-card">
            <h3>Demo Intent</h3>
            <p>
              This interface is optimized for executive demos and architectural storytelling,
              showcasing senior-level engineering quality and product thinking.
            </p>
          </div>
        </aside>
      </section>

      <footer className="site-footer">
        <p>© 2023 Bertrand Amobi.</p>
        <p>For demonstration purposes only. <a href="https://github.com/bertrandamobi/AI-Workflow-Engine" target="_blank" rel="noreferrer">View Source Code.</a></p>
      </footer>
    </main>
  );
}


================================================
FILE: frontend/src/app/styles.css
================================================
:root {
  color-scheme: dark;
}

* {
  box-sizing: border-box;
}

body {
  margin: 0;
  min-height: 100vh;
  background: radial-gradient(circle at 20% 20%, #341426, #0f0f13 45%),
    radial-gradient(circle at 80% 70%, #20403c, transparent 30%),
    linear-gradient(140deg, #1e0f2b, #0f1119 70%);
  color: #f8f4f0;
  font-family: Inter, Segoe UI, system-ui, -apple-system, sans-serif;
}

.page-shell {
  position: relative;
  overflow: hidden;
  padding: 2.5rem 1.25rem 1.5rem;
  max-width: 1180px;
  margin: 0 auto;
}

.aurora {
  position: absolute;
  filter: blur(60px);
  border-radius: 999px;
  opacity: 0.5;
  animation: drift 15s ease-in-out infinite alternate;
  z-index: -2;
}

.aurora-a {
  width: 300px;
  height: 300px;
  background: #f25f4c;
  top: -120px;
  left: -100px;
}

.aurora-b {
  width: 340px;
  height: 340px;
  background: #59c3c3;
  right: -120px;
  bottom: 80px;
  animation-delay: 1.5s;
}

.noise-overlay {
  position: absolute;
  inset: 0;
  pointer-events: none;
  z-index: -1;
  opacity: 0.12;
  background-image: radial-gradient(circle, #f6bd6022 1px, transparent 1px);
  background-size: 3px 3px;
}

.hero h1 {
  font-size: clamp(2rem, 4vw, 3.5rem);
  margin: 0.4rem 0;
}

.eyebrow {
  color: #f6bd60;
  letter-spacing: 0.1em;
  text-transform: uppercase;
  font-size: 0.78rem;
}

.hero-copy {
  max-width: 68ch;
  color: #e6ddd5;
}

.hero-metrics {
  display: grid;
  grid-template-columns: repeat(auto-fit, minmax(130px, 1fr));
  gap: 1rem;
  margin-top: 1.2rem;
}

.hero-metrics article {
  background: #ffffff0f;
  border: 1px solid #ffffff24;
  border-radius: 14px;
  padding: 0.9rem;
  backdrop-filter: blur(6px);
}

.hero-metrics span {
  font-weight: 700;
  color: #ffdca8;
}

.content-grid {
  margin-top: 2rem;
  display: grid;
  grid-template-columns: 1.6fr 1fr;
  gap: 1.2rem;
}

.panel {
  border-radius: 18px;
  padding: 1.25rem;
}

.elevated {
  background: linear-gradient(160deg, #31142f, #182227);
  border: 1px solid #ffffff1a;
  box-shadow: 0 30px 80px #0000004f;
}

.glass {
  background: #ffffff0f;
  border: 1px solid #ffffff2b;
  backdrop-filter: blur(8px);
}

.panel-copy {
  color: #eadfd7;
}

ul {
  margin: 0;
  padding-left: 1.2rem;
  display: grid;
  gap: 0.65rem;
}

.pulse-card {
  margin-top: 1rem;
  padding: 1rem;
  border-radius: 14px;
  background: linear-gradient(120deg, #f25f4c2d, #59c3c32e);
  border: 1px solid #ffffff24;
  animation: pulse 4s ease-in-out infinite;
}

form {
  display: grid;
  gap: 0.8rem;
}

input,
textarea,
button {
  border-radius: 12px;
  border: 1px solid #ffffff2d;
  font: inherit;
  padding: 0.75rem 0.8rem;
}

input,
textarea {
  background: #0e1118a8;
  color: #f5f2ee;
}

button {
  background: linear-gradient(110deg, #f25f4c, #f6bd60);
  color: #141414;
  font-weight: 700;
  cursor: pointer;
  transition: transform 160ms ease, filter 160ms ease;
}

button:hover {
  transform: translateY(-1px);
  filter: brightness(1.07);
}

.site-footer {
  margin-top: 2rem;
  border-top: 1px solid #ffffff2a;
  padding-top: 1rem;
  color: #e0d8d0;
  text-align: center;
}

.site-footer p {
  margin: 0.3rem 0;
}

.site-footer a {
  color: #ffdca8;
}

@media (max-width: 900px) {
  .content-grid {
    grid-template-columns: 1fr;
  }
}

@keyframes pulse {
  0%, 100% { transform: scale(1); }
  50% { transform: scale(1.01); }
}

@keyframes drift {
  from { transform: translateY(0) translateX(0); }
  to { transform: translateY(30px) translateX(18px); }
}


================================================
FILE: frontend/src/components/WorkflowForm.tsx
================================================
"use client";

import { FormEvent, useState } from "react";
import { runMarketAnalysis } from "../lib/api";
import { WorkflowResponse } from "../types/workflow";

export default function WorkflowForm() {
  const [company, setCompany] = useState("NVIDIA");
  const [objectives, setObjectives] = useState("Competitive positioning, product strategy");
  const [result, setResult] = useState<WorkflowResponse | null>(null);
  const [loading, setLoading] = useState(false);

  const onSubmit = async (event: FormEvent) => {
    event.preventDefault();
    setLoading(true);
    setResult(null);
    try {
      const response = await runMarketAnalysis({
        company,
        objectives: objectives.split(",").map((item) => item.trim()),
      });
      setResult(response);
    } finally {
      setLoading(false);
    }
  };

  return (
    <div>
      <form onSubmit={onSubmit}>
        <input value={company} onChange={(event) => setCompany(event.target.value)} placeholder="Company" />
        <textarea
          value={objectives}
          onChange={(event) => setObjectives(event.target.value)}
          placeholder="Comma-separated objectives"
        />
        <button type="submit" disabled={loading}>{loading ? "Running..." : "Run Workflow"}</button>
      </form>

      {result && (
        <section>
          <h2>Workflow Result</h2>
          <p><strong>ID:</strong> {result.workflow_id}</p>
          <p><strong>Status:</strong> {result.status}</p>
          <h3>Summary</h3>
          <p>{result.summary}</p>
          <h3>Recommendations</h3>
          <p>{result.artifacts.recommendations}</p>
        </section>
      )}
    </div>
  );
}


================================================
FILE: frontend/src/lib/api.ts
================================================
import { WorkflowRequest, WorkflowResponse } from "../types/workflow";

const API_BASE_URL = process.env.NEXT_PUBLIC_API_BASE_URL ?? "http://localhost:8000";

export async function runMarketAnalysis(payload: WorkflowRequest): Promise<WorkflowResponse> {
  const response = await fetch(`${API_BASE_URL}/workflows/market-analysis`, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(payload),
  });

  if (!response.ok) {
    throw new Error(`Request failed: ${response.status}`);
  }
  return response.json();
}


================================================
FILE: frontend/src/types/workflow.ts
================================================
export interface WorkflowRequest {
  company: string;
  objectives: string[];
}

export interface WorkflowResponse {
  workflow_id: string;
  status: string;
  summary: string;
  artifacts: {
    research_notes: string;
    recommendations: string;
  };
}


================================================
FILE: frontend/tsconfig.json
================================================
{
  "compilerOptions": {
    "target": "ES2022",
    "lib": ["DOM", "DOM.Iterable", "ES2022"],
    "allowJs": false,
    "skipLibCheck": true,
    "strict": true,
    "noEmit": true,
    "module": "ESNext",
    "moduleResolution": "Bundler",
    "resolveJsonModule": true,
    "isolatedModules": true,
    "jsx": "preserve",
    "incremental": true
  },
  "include": ["src"]
}