MongoStory is a cloud-native platform designed to empower content creators, editors, and publishers with AI-driven tools for content management and distribution. Built with Next.js 15 and MongoDB, the platform offers a comprehensive suite of features for content creation, analysis, translation, and optimization.

• Content Creation & Editing: Rich markdown editor for creating and editing content
• Content Versioning: Track revisions and compare different versions of content
• Publishing Workflow: Draft, review, and publish content with status tracking
• Content Organization: Semantic clustering and intelligent categorization

• Content Analysis: Analyze content quality, readability, and structure
• SEO Optimization: AI-generated recommendations for improving search visibility
• Emotional Impact Analysis: Understand the emotional tone and impact of content
• Topic Analysis: Identify main topics and suggest related content areas
• Automatic Translation: Translate content to multiple languages with AI
• Social Media Post Generation: Create platform-specific social media posts from content

• Performance Metrics: Track views, engagement, and user behavior
• Content Intelligence: AI-driven insights about content relationships and gaps
• Semantic Search: Find content based on meaning rather than just keywords
• Topic Distribution: Visualize content distribution across topics
• Content Clusters: Automatically group related content by semantic similarity

• Framework: Next.js 15 with App Router
• UI Components: Custom components built with shadcn/ui
• State Management: React Context API for global state
• Authentication: JWT-based authentication system

• API Routes: Next.js API routes for server-side functionality
• Database: MongoDB for flexible document storage
• Vector Search: MongoDB Atlas Vector Search for semantic content operations
• AI Integration: Integration with AI models via AI SDK - xAI (Grok)

• Content Generation: AI-powered content creation and suggestions
• Analysis: Multi-faceted content analysis (SEO, quality, emotion, topics)
• Vector Embeddings: Content embeddings for semantic search and clustering - VoyageAI embeddings
• Translation: Multilingual content support with AI translation

MongoStory leverages MongoDB's document model for flexible content storage and its vector search capabilities for advanced content operations:

• Collections Structure:

  • content: Stores all content items with embedded analysis and translations
  • users: User accounts and authentication information
  • analytics: Content performance metrics
  • clusters: AI-generated content clusters
  • socialMediaPosts: Generated social media content

• Vector Search: Uses MongoDB Atlas Vector Search for semantic operations:

  • Content similarity detection
  • Semantic search functionality
  • Automatic content clustering

• Node.js 18+ and npm/yarn
• MongoDB Atlas account
• AI API keys:
• • Grok AI API key
• • Voyage AI API Key.

1. Clone the repository
2. Install dependencies with

npm install
## npm install --force

3. Configure environment variables
4. Run the development server with npm run dev

• MONGODB_URI: Connection string for MongoDB Atlas
• JWT_SECRET: Secret for JWT authentication

openssl rand -base64 32

• XAI_API_KEY: API key for AI services
• VOYAGE_API_KEY: API key for vector embeddings
• NEXT_PUBLIC_APP_URL: The main domain of the app (eg. http://localhost:3000).

Set a VOYAGE_API_KEY - Value + Secret on the triggers app.

Create the following Atlas Trigger to be placed on insert for mongostory.content collection:

exports = async function(changeEvent) {
  // Get the full document that triggered the event
  const fullDocument = changeEvent.fullDocument;

  // Extract the text field that needs to be embedded
  // Change 'text' to whatever field contains the content you want to embed
  const textToEmbed = fullDocument.analysis.summary;

  if (!textToEmbed) {
    console.log("No text field found in the document");
    return;
  }

  try {
    // Connect to your MongoDB cluster
    const collection = context.services.get("ILCluster").db("mongostory").collection("content");

    // Call Voyage AI API to generate embeddings
    const response = await context.http.post({
      url: "https://api.voyageai.com/v1/embeddings",
      headers: {
        "Authorization": [`Bearer ${context.values.get("VOYAGE_API_KEY")}`],
        "Content-Type": ["application/json"]
      },
      body: JSON.stringify({
        "input": [textToEmbed],
        "model": "voyage-3",
        // Optional parameters based on your needs
        "input_type": "document", // Use "query" for search queries, "document" for content
      })
    });

    // Parse the response
    const responseData = EJSON.parse(response.body.text());

    // Get the embedding from the response
    const embedding = responseData.data[0].embedding;

    // Update the document with the embedding
    await collection.updateOne(
      { _id: fullDocument._id},
      { $set: { embedding: embedding } }
    );

    console.log(`Successfully added embedding to document ${fullDocument._id}`);

    // Optional: Create a vector search index if it doesn't exist
    // Note: This should ideally be done once, not in every trigger execution
    // This is just for demonstration purposes

    return { status: "success" };

  } catch (error) {
    console.error("Error in Voyage AI embedding trigger:", error);
    return { status: "error", message: error.message };
  }
};

Create a vector search index (named vector_index) on content collection:

{
  "fields": [
    {
      "type": "vector",
      "path": "embedding",
      "numDimensions": 1024,
      "similarity": "cosine"
    }
  ]
}

The central hub providing access to all platform features with overview metrics and quick actions.

Interface for creating, editing, and managing content with AI assistance.

Detailed metrics on content performance, user engagement, and distribution channels.

AI-powered insights into content relationships, gaps, and optimization opportunities.

Visual representation of the database structure and relationships.

• Workflow Automation: Advanced publishing workflows with approval processes
• Personalization Engine: Content personalization based on user behavior
• Advanced Analytics: Predictive analytics for content performance
• Multi-channel Distribution: Expanded distribution options beyond social media
• Collaboration Tools: Enhanced team collaboration features

Contributions to MongoStory are welcome! Please refer to the CONTRIBUTING.md file for guidelines.

This project is licensed under the MIT License - see the LICENSE file for details.