Merge branch 'main' into neo4j-starter-kit

.github/workflows/build.yml

@@ -9,6 +9,8 @@ on:
       - ready_for_review
     paths:
       - "**.go"
+      - "**.mod"
+      - "**.sum"
       - "**.ts"
       - "**.json"
       - .github/workflows/*

dgraph-101/api-as/package.json

@@ -9,7 +9,7 @@
     "pretty:check": "prettier --check ."
   },
   "dependencies": {
-    "@hypermode/modus-sdk-as": "^0.15.0",
+    "@hypermode/modus-sdk-as": "^0.16.0",
     "assemblyscript-json": "^1.1.0",
     "json-as": "0.9.26"
   },

dgraph-modus-example/README.md

@@ -0,0 +1,138 @@
+# Movie Recommendation App
+
+This project demonstrates the power of **Modus** and **Dgraph** by integrating a graph database with
+a large language model (LLM) to provide AI-driven movie recommendations. The app consists of a
+backend (powered by Modus and Dgraph) and a frontend (built with Next.js and tailwind).
+
+## Why Use Modus with Dgraph?
+
+**Modus** simplifies the integration of internal data (from sources like Dgraph) with cutting-edge
+LLMs. Here's why this combination is powerful:
+
+- **Rich Data Queries**: Dgraph enables complex queries across interconnected datasets like movies,
+  genres, actors, and directors. This makes it ideal for graph-based recommendations.
+- **Seamless AI Integration**: Modus abstracts the complexity of using LLMs, allowing you to feed
+  internal data into models and retrieve actionable insights tailored to your data.
+- **Rapid Prototyping**: With Modus, you can quickly build and test serverless functions that
+  combine graph-based querying and AI recommendations.
+- **Customizability**: Dynamically construct prompts for LLMs using Dgraph's query results to
+  generate personalized outputs, such as movie recommendations.
+
+---
+
+## Backend Setup
+
+The backend is written in Go using **Modus** and **Dgraph**.
+
+### Features
+
+- Functions powered by Modus are integrated with a read-only Dgraph database and an LLM from Hugging
+  Face.
+- Provides a GraphQL endpoint for querying data locally.
+- Demonstrates how to use Modus to create powerful, serverless APIs that leverage Dgraph's
+  capabilities and LLM-driven insights.
+
+### Prerequisites
+
+- Install **Modus CLI**: Follow the
+  [Modus installation guide](https://docs.hypermode.com/modus/installation).
+- Ensure **Dgraph** is available or connected to the backend (the project uses a read-only Dgraph
+  database).
+
+### Running Backend Locally
+
+Move into the `backend` directory:
+
+    ```bash
+    cd backend
+    ```
+
+Install dependencies (if required):
+
+    ```bash
+    go mod tidy
+    ```
+
+Start the backend using Modus:
+
+    ```bash
+    modus dev
+    ```
+
+Access the local endpoint:
+
+    - GraphQL API Explorer: `http://localhost:8686/graphql`
+    - Query APIs or explore the auto-generated schema directly in the explorer.
+
+Key Notes - Functions are defined in Go and connected to a read-only Dgraph database. - The backend
+integrates with an LLM to generate movie recommendations dynamically. - Check the Modus logs to see
+which endpoints and functions are exposed.
+
+---
+
+## Frontend Setup
+
+The frontend is built with **Next.js** and **Tailwind**.
+
+### Frontend Features
+
+- Search for movies by title or related fields.
+- View movie details, including release year, genres, and starring actors.
+- Get AI-driven recommendations for similar movies.
+
+### Running Frontend Locally
+
+Move into the frontend directory:
+
+    ```bash
+    cd frontend
+    ```
+
+Install dependencies:
+
+    ```bash
+    npm install
+    ```
+
+Start the development server:
+
+    ```bash
+    npm run dev
+    ```
+
+4.  Open the app in your browser: `http://localhost:3000`.
+
+### Important
+
+- Ensure the backend is running locally for the frontend to fetch data and generate recommendations.
+- The backend API endpoints are integrated into the frontend through `actions.ts`.
+
+---
+
+## How It Works
+
+1. Backend:
+
+   - Queries Dgraph for movie data using GraphQL-like queries.
+   - Dynamically generates prompts for the LLM to fetch recommendations.
+   - Returns combined movie details and recommendations as a JSON response.
+
+2. Frontend:
+
+   - Fetches data from the backend's GraphQL endpoints.
+   - Displays movie details and AI-generated recommendations.
+   - Provides a clean, user-friendly interface for exploring movies and related data.
+
+---
+
+## Learning More
+
+- Modus Documentation:
+  [https://docs.hypermode.com/modus/overview](https://docs.hypermode.com/modus/overview)
+- Dgraph Documentation: [https://dgraph.io/docs](https://dgraph.io/docs)
+- Next.js Documentation: [https://nextjs.org/docs](https://nextjs.org/docs)
+
+---
+
+This project is an excellent starting point for understanding how to combine the strengths of
+Dgraph, Modus, and LLMs to create powerful, data-driven applications. 🚀