How to Build a Real-Time Equity Research Dashboard using Kite MCP & Yahoo Finance (on Emergent)

In this tutorial, you’ll build Market Snapshot - a live market dashboard that lets you:

• Search and add stocks to your watchlist directly using Kite MCP

• View NIFTY50 and BANKNIFTY index performance in real time

• Track your watchlist stocks with live price, % change, volume, day high/low

• View a 5-minute intraday line chart for each stock

• See matched news headlines for each selected stock along with quick sentiment tags

Everything is built end-to-end on Emergent using plain English prompts - no manual coding, chart wiring, API setup, or backend logic required.

In Simple Terms

• Kite MCP gives you live market and watchlist data.

• Yahoo Finance provides news headlines that mention the stocks in your watchlist.

• Emergent handles the rest:
  
  

  • Connecting to the APIs

  • Refreshing data automatically

  • Displaying prices and intraday charts

  • Running a simple sentiment analysis on headlines

  • Presenting everything in a clean, dark trading dashboard UI

What You’ll Have by the End

Your dashboard will include:

How to configure Kite MCP

Before you submit the prompt to build the Market Snapshot app, you need to connect your Zerodha account’s market data feed to Emergent using Kite MCP. This allows Emergent to fetch live prices, volumes, and index values directly from your brokerage account.

Emergent provides an MCP (Model Control Panel) interface where you can add external data sources.
We’ll use this to plug in Kite MCP.

Step 1: Open MCP Configuration in Emergent

1. Go to Emergent

2. Open the Advanced Controls menu (top-right developer tools area)

3. Click on MCP

4. Select MCP Tools

5. Click Add New MCP Server

This will open the configuration modal.

You’ll see a form like this (as shown in the screenshot):

• MCP Name

• Description

• JSON Configuration

Step 2: Enter MCP Server Details

Fill the fields as follows:

JSON Configuration (Copy & Paste)

{

  "mcpServers": {

    "kite": {

      "args": [

        "mcp-remote",

        "https://mcp.kite.trade/mcp"

      ],

      "command": "npx"

    }

  }

}

This configuration is provided directly by Zerodha:
https://zerodha.com/z-connect/featured/connect-your-zerodha-account-to-ai-assistants-with-kite-mcp

Step 3: Save & Verify

Click Verify and Save.

Emergent will now connect to the Kite MCP remote server.
If prompted at any time during app usage, you will sign in with your Zerodha login to authorize data access.

Once verified, Kite MCP is now available to use in any app you build on Emergent.

Building on Emergent

Once you're on Emergent, the entire app begins with a single detailed prompt. This prompt tells the agent what to build, how the data should flow, what APIs to connect to, and how the dashboard should look. Emergent will handle the backend logic, API requests, authentication, charting, and UI layout automatically.

Here is the exact prompt used to build the Market Snapshot dashboard:

Build a real-time web dashboard app called “Market Snapshot” that integrates data from Kite MCP and Yahoo Finance.

---

Core Features

1. Watchlist + Stock Search

- Allow users to search for NSE/BSE symbols using Kite MCP.

- Enable “Add to Watchlist” functionality.

- Display each watchlist stock with:

  - Last traded price (LTP)

  - Percent change

  - Volume

  - Day high and day low

- Auto-refresh data every 30 seconds.

2. Index Performance

- Fetch and display real-time values for NIFTY50 and BANKNIFTY.

- Show absolute and percentage change with up/down indicators.

3. Intraday Chart

- For each stock in the watchlist, show a 5-minute interval intraday price line chart.

- Update charts on refresh.

4. News Integration

- Fetch the latest news related to each stock using Yahoo Finance News API.

- Match news to the stock symbol.

- Run a basic sentiment check to label each headline as:

  - Positive

  - Neutral

  - Negative

- Display sentiment badge alongside each headline.

---

User Experience & Design

- Use a clean, dark-mode trading dashboard interface.

- Show smooth color transitions when price goes up or down.

- Layout sections clearly:

  - Top: Index cards

  - Middle: Watchlist table

  - Below each stock: Intraday chart + latest news

- Refresh button + auto-refresh every 30 seconds.

Once you submit this prompt, Emergent will read it and begin to assemble the app. It will then ask a few clarifying questions so it knows how to configure your Kite MCP and news data

Explaining the Prompt (In Simple Terms)

This prompt tells Emergent what data the app should use, how it should display that data, and how users should interact with the dashboard. The first part of the prompt explains that the app will connect to Kite MCP, which is the live market data stream from Zerodha. This is what allows the dashboard to show real-time prices for stocks and indices like NIFTY50 and BANKNIFTY. The prompt also specifies that users should be able to search for stocks and add them to a personal watchlist, rather than relying on predefined lists or holdings. This makes the dashboard personalized and interactive.

Next, the prompt introduces Yahoo Finance as the source for news. For every stock a user adds to the watchlist, the app will fetch the latest related headlines. It will also run a simple sentiment check to label each news item as Positive, Neutral, or Negative. This helps users quickly understand the market’s mood around a stock without reading every article in detail. The news appears right below each stock - keeping price and context together in one place.

The prompt also specifies that each stock should display an intraday chart using 5-minute interval data. This lets users see how the stock has been trending throughout the trading day - whether the price has been moving steadily up, sliding downward, or oscillating. It gives a visual snapshot of momentum without overwhelming the user with complex technical indicators.

Finally, the prompt defines the overall experience: the dashboard should use a clean, dark-mode interface with smooth updates as prices and sentiment change. The layout is simple and structured - index performance at the top, your watchlist in the center, and charts and news expanding beneath each stock. The dashboard updates automatically every 30 seconds, so the user can leave it open during market hours and rely on it for ongoing visibility.

In short, the prompt tells Emergent to combine live market data, relevant news, and intuitive visualization into a dashboard that is personal, real-time, and easy to navigate - all without requiring the user to write any code.

Authenticate Your Kite MCP Connection

Once you submit the prompt, Emergent will detect that the app requires live market data and will prompt you to authenticate your Zerodha account through Kite MCP.

You will see a message like this:

🔐 Kite MCP Authentication Required

To fetch your market data, please authenticate your Zerodha account by clicking the link below:

👉 Login to Kite

Or copy this URL to your browser:

https://kite.zerodha.com/connect/login?api_key=kitemcp&v=3&redirect_params=session_id%3D...

📋 What happens next:

1. Click the login link above

2. Enter your Zerodha User ID and Password

3. Complete 2FA (PIN / OTP)

4. Authorize the connection when prompted

5. Return to Emergent - your MCP session will now be active

What This Step Does

This step securely links your dashboard to your real market feed.

No passwords are stored in Emergent.

Authentication happens directly on Zerodha’s server.

Once authenticated, Emergent can:

Fetch live prices for watchlist stocks

Retrieve index values (NIFTY50, BANKNIFTY)

Search ticker names to add to your watchlist

Agent’s Questions & Answers

After you authenticate with Kite MCP, the Emergent agent will ask a few clarifying questions before it begins building the dashboard. These questions ensure that the app behaves the way you expect - especially for watchlist handling, historical data, and news sentiment.

Here are the questions the agent asks, followed by the answers used:

1. Watchlist Behavior

Agent asks:

For your watchlist, should I:
a. Fetch your actual holdings from Zerodha
b. Allow you to manually select/add stocks to watch
c. Use a predefined list of popular stocks

Answer:
b. Allow me to manually select and add stocks to watch

2. Yahoo Finance Data Source

Agent asks:

I'll use the yfinance library for news.
Should I also fetch real-time prices from Yahoo as a backup to Kite?

Answer:
No - only use yfinance for news. All price data should come from Kite.

3. Sentiment Model

Agent asks:

Which sentiment model should I use?
a. Basic TextBlob
b. VADER (better for financial tone)

Answer:
a. Use basic TextBlob sentiment analysis

4. Historical Data for Line Chart

Agent asks:

What timeframe should the intraday chart display?

Answer:
Show 1 Day (5-minute interval candlestick-derived line chart)

5. News Matching Method

Agent asks:

Should news be matched using keyword/company names or by stock symbol?

Answer:
Fetch news for each stock symbol separately

How to Deploy Your App for Everyone to Use

Once the agent finishes building the Market Snapshot dashboard and the preview looks good, you can deploy the app so that it stays live and accessible anytime. Deployment ensures the app runs on dedicated servers and does not go to sleep. This keeps the data streaming continuously and makes the dashboard shareable with others.

To deploy, go to the top-right corner of the Emergent interface and click Deploy. Before deployment begins, run the Pre-deployment Health Check. This scan confirms that the app’s data connections, UI elements, and backend processes are stable. Once the health check completes successfully, click Start Deployment.

The agent will automatically handle the entire deployment process. It will package the app, migrate any required data layers, securely export all API keys and tokens, and start the production environment. After deployment, another automated health check runs to verify that everything is working as expected.

When deployment is finished, you will receive a live link that keeps the dashboard active around the clock. You can share this link with others or connect a custom domain for a fully branded experience. Keeping the app deployed costs 50 credits per month, which maintains the servers and ensures the dashboard remains always-on and real-time.

How You Can Improve the App

Once your Market Snapshot dashboard is running, you can extend it using small, targeted prompts. Emergent will modify the app automatically without you rewriting code. Here are a few useful enhancements you can add:

1. Add Alerts for Price Movements

Allow the app to notify you when a stock crosses a certain price or percentage change.

Prompt to use:

Add a price alert system to the Market Snapshot app. Allow users to set alert conditions for each watchlist stock (like price crossing a value or percent change threshold). Notify the user inside the dashboard when alerts trigger.

2. Add Support for F&O Stocks with OI Data

Include Open Interest and OI change to monitor futures activity.

Prompt to use:

Extend the Market Snapshot watchlist to support F&O stocks. Fetch and display Open Interest and Change in OI from Kite MCP alongside price and volume.

3. Add Technical Indicators to Charts

Show indicators such as Moving Averages or RSI on the intraday chart.

Prompt to use:

Enhance the intraday chart by adding 5-day and 20-day moving averages and RSI. Display the indicators as overlays and place them in the expanded chart view for each stock.

4. Add Earnings Dates & Corporate Events

Show upcoming results announcements or dividends for each company.

Prompt to use:

Fetch and display upcoming earnings dates, dividends, and corporate event information for each stock in the watchlist. Show this data below the stock’s news section.

5. Add Portfolio P&L View (Optional)

If you want to track your actual holdings, not just watchlist stocks.

Prompt to use:

Add a Portfolio tab that fetches my actual holdings from Kite MCP and displays quantity, average buy price, current value, and total P&L in a clean table.

Each of these enhancements works as a small, natural-language update. Just paste the prompt into Emergent and the agent will adjust the app - no code changes required.