competitor-analysis

Competitive pricing intelligence (TinyFish cookbook)

Package: competitor-analysis (see package.json)

Live demo: https://competitor-priceanalysis.vercel.app/

A competitive pricing intelligence app for tracking many competitors at once. It follows Source → Extract → Present: OpenRouter helps discover pricing URLs, Tinyfish Agent (via @tiny-fish/sdk) scrapes and extracts structured JSON from pricing pages in parallel, and OpenRouter analyzes the results for the dashboard.

Status: Working

---

Stack

Piece	Role
Next.js 16 (App Router)	UI and API routes
@tiny-fish/sdk	Tinyfish Agent streaming runs (client.agent.stream) in /api/scrape-pricing
OpenRouter	URL hints, pricing analysis (/api/generate-urls, /api/analyze-pricing)
React 19, Tailwind, shadcn/ui	Frontend

---

How Tinyfish Agent is wired

Scraping is implemented in app/api/scrape-pricing/route.ts: for each competitor URL it calls new TinyFish({ apiKey }).agent.stream({ url, goal, browser_profile: 'lite' }), forwards step and streamingUrl events to the browser over SSE, and on COMPLETE maps resultJson into the app’s CompetitorPricing schema. Environment variable: TINYFISH_API_KEY.

---

Demo

[Demo video/screenshot to be added]

---

Quick start

npm install
export TINYFISH_API_KEY=your_key
export OPENROUTER_API_KEY=your_key
npm run dev

Or use a .env.local file (see .env.local.example).

---

How to Run

Prerequisites

• Node.js 18+
• Tinyfish API key (set TINYFISH_API_KEY)

Setup

1. Clone the repository:

git clone https://github.com/tinyfish-io/TinyFish-cookbook
cd TinyFish-cookbook/competitor-analysis

2. Install dependencies:

npm install

3. Create .env.local file:

TINYFISH_API_KEY=xxx          # Browser automation
OPENROUTER_API_KEY=xxx    # AI URL generation + pricing analysis

4. Run the development server:

npm run dev

5. Open http://localhost:3000 in your browser

---

Architecture Diagram

flowchart TB
    subgraph Input[USER INPUT]
        Step1[Step 1: Baseline Pricing]
        Step2[Step 2: Competitor List + Detail Level]
    end
    subgraph Phase1[PHASE 1: URL GENERATION]
        AI1[OpenRouter AI]
        URLs[Generated Pricing URLs]
    end
    subgraph Phase2[PHASE 2: PARALLEL SCRAPING]
        A1[Agent 1]
        A2[Agent 2]
        A3[Agent 3]
        A15[Agent 15...]
    end
    subgraph Phase3[PHASE 3: AI ANALYSIS]
        AI2[OpenRouter AI]
        Analysis[Strategic Insights]
    end
    subgraph Output[RESULTS]
        Dashboard[4-Tab Dashboard]
        Export[CSV/JSON Export]
    end
    Step1 --> Step2
    Step2 --> AI1
    AI1 --> URLs
    URLs --> A1
    URLs --> A2
    URLs --> A3
    URLs --> A15
    A1 --> AI2
    A2 --> AI2
    A3 --> AI2
    A15 --> AI2
    AI2 --> Analysis
    Analysis --> Dashboard
    Dashboard --> Export
    A1 -.-> Dashboard
    A2 -.-> Dashboard
    A3 -.-> Dashboard
    A15 -.-> Dashboard

Loading

sequenceDiagram
    participant U as User
    participant FE as Frontend
    participant Ctx as PricingContext
    participant API1 as /api/generate-urls
    participant API2 as /api/scrape-pricing
    participant API3 as /api/analyze-pricing
    participant AI as OpenRouter AI
    participant M as Tinyfish Agent Runs
    U->>FE: Enter baseline pricing
    FE->>Ctx: Store baseline
    U->>FE: Add 10-15 competitors
    FE->>Ctx: Store competitors
    FE->>API1: Generate missing URLs
    API1->>AI: Generate pricing URLs
    AI-->>API1: URLs with confidence
    API1-->>FE: Enriched competitors
    U->>FE: Start scraping
    FE->>API2: POST /api/scrape-pricing
    API2->>M: Launch parallel agents
    M-->>API2: Stream live URLs
    API2-->>FE: Forward streaming URLs
    FE-->>U: Show live competitor grid
    M-->>API2: Extract pricing data
    API2-->>FE: Stream results
    FE-->>U: Update dashboard
    API2->>API3: Trigger analysis
    API3->>AI: Analyze pricing structures
    AI-->>API3: Insights + recommendations
    API3-->>FE: Analysis complete
    FE-->>U: Display insights

Loading

classDiagram
    class BaselinePricing {
        +string companyName
        +string pricingModel
        +string unitType
        +number pricePerUnit
        +string currency
    }
    class Competitor {
        +string id
        +string name
        +string url
        +string generatedUrl
        +string urlConfidence
    }
    class CompetitorPricing {
        +string company
        +string url
        +string pricingModel
        +string primaryUnit
        +string unitDefinition
        +PricingTier[] tiers
        +string additionalNotes
        +datetime scrapedAt
    }
    class PricingTier {
        +string name
        +number price
        +string billingPeriod
        +string unit
        +string includedUnits
        +string overagePrice
    }
    class Analysis {
        +string[] insights
        +string[] recommendations
        +Record pricingModelBreakdown
        +Record normalizedPrices
        +number yourPosition
    }
    class ScrapingStatus {
        +string status
        +string streamingUrl
        +string[] steps
        +CompetitorPricing data
        +string error
    }

Loading