MonitorMamdani.com Case Study: Virlo API + Kalshi + Polymarket + Firecrawl

MonitorMamdani.com was built around a simple idea:

Prediction markets tell you what people think will happen.
Viral attention and news explain why those beliefs are changing.

Political outcomes today are shaped in real time by short-form video, social narratives, and rapidly shifting media cycles. Odds can move sharply before traditional analysis even catches up.

To solve this, MonitorMamdani.com combines real-time prediction market data from Kalshi and Polymarket with viral attention data from the Virlo API and automated news ingestion via Firecrawl, all in a single monitoring dashboard.

The problem with standalone prediction markets

Platforms like Kalshi and Polymarket are extremely good at aggregating sentiment into probabilities. But they intentionally stop short of explaining what’s driving those probabilities.

If you’re watching a market move from 42% to 51%, the obvious next questions are:

• What narrative changed?

• Did something go viral?

• Was there a specific article, clip, or talking point that triggered the shift?

• Is the move supported by attention momentum or just short-term noise?

Without external context, those answers require manual work across multiple platforms.

The MonitorMamdani.com solution

MonitorMamdani.com was designed to layer context directly on top of prediction market odds by combining three data sources:

1. Prediction market odds from

   • https://kalshi.com

   • https://polymarket.com

2. Viral short-form attention data from the Virlo API

   • https://virlo.ai

   • https://dev.virlo.ai

3. Structured news ingestion using Firecrawl

   • https://firecrawl.dev

This allows users to see not just where the market is, but why it’s moving.

Layer 1: Prediction market data (Kalshi + Polymarket)

Kalshi

MonitorMamdani.com pulls market data from Kalshi’s public API, including:

• Active and resolved political markets

• Market titles, tickers, and metadata

• Current implied probabilities

Kalshi markets are discovered and refreshed using endpoints documented at:
https://docs.kalshi.com/api-reference/market/get-markets

Polymarket

Polymarket data is sourced from the Gamma Markets API, which provides a structured, read-only index of Polymarket events and markets:

• https://gamma-api.polymarket.com/markets

• https://gamma-api.polymarket.com/events

This allows MonitorMamdani.com to:

• Discover relevant political markets

• Track odds changes over time

• Normalize Polymarket data alongside Kalshi in a shared format

Both Kalshi and Polymarket odds are displayed side-by-side, making divergence and convergence immediately visible.

Layer 2: Viral context via the Virlo API

Prediction markets move fastest when attention compounds. That attention increasingly originates from short-form video platforms like TikTok, YouTube Shorts, and Instagram Reels.

MonitorMamdani.com uses the Virlo API, specifically Virlo’s Orbit social listening system, to track:

• Viral videos mentioning Zohran Mamdani

• Topic clusters forming around policies, controversies, or messaging

• Velocity and breakout patterns across platforms

Virlo’s Orbit searches are keyword-driven and refresh continuously, allowing MonitorMamdani.com to align market movements with real-time attention shifts.

Virlo API documentation:
https://dev.virlo.ai

Virlo platform overview:
https://virlo.ai

This transforms prediction market monitoring from “price watching” into narrative monitoring.

Layer 3: News ingestion with Firecrawl

Social attention doesn’t exist in isolation. Viral moments are often triggered or amplified by traditional reporting.

To ingest and structure news reliably, MonitorMamdani.com uses Firecrawl, a developer-first web scraping and content extraction platform.

Firecrawl enables:

• Crawling and extracting dynamic news pages

• Converting articles into clean Markdown or structured JSON

• Avoiding brittle, site-specific scrapers

Firecrawl homepage:
https://firecrawl.dev

By using Firecrawl, MonitorMamdani.com can:

• Pull the exact articles driving discussion

• Store clean, readable content

• Associate coverage with specific market movements

How the system works end-to-end

Step 1: Market discovery and refresh

• Kalshi markets are fetched via their markets API

• Polymarket markets are fetched via Gamma endpoints

• Odds are normalized into a shared schema

Step 2: Movement detection

The system flags:

• Significant probability changes

• Rapid rate-of-change

• Disagreements between Kalshi and Polymarket

Step 3: Context hydration with Virlo

When a market moves:

• A Virlo Orbit query is triggered

• Viral videos, topics, and velocity metrics are retrieved

• Attention data is aligned to the movement window

Step 4: News hydration with Firecrawl

Relevant articles are:

• Crawled

• Cleaned

• Stored alongside market and viral data

Step 5: Unified dashboard display

Each monitored outcome shows:

• Kalshi odds

• Polymarket odds

• Viral attention context (Virlo)

• Relevant news coverage (Firecrawl)

Why this approach matters

Faster interpretation

Markets often move before commentary stabilizes. Viral data explains why before consensus forms.

Better signal vs noise

Attention velocity and topic clustering help distinguish:

• Short-lived outrage

• Sustained narrative shifts

One monitoring surface

Instead of bouncing between:

• Prediction markets

• Social platforms

• News sites

Everything lives in one place.

Key takeaways

MonitorMamdani.com demonstrates how modern political intelligence benefits from stacking probability with attention.

By combining:

• https://kalshi.com

• https://polymarket.com

• https://virlo.ai

• https://firecrawl.dev

the platform turns raw odds into explainable market movements.

Prediction markets answer what people believe.
Virlo and Firecrawl explain why they believe it