# AI Crawler Guidelines for Cresco
# This file provides additional context for AI crawlers and language models

## About Cresco

Cresco is an AI-powered Korean stock market analysis platform that provides real-time financial data analysis for KOSPI and KOSDAQ companies.

## Product Information

- **Name**: Cresco
- **Category**: Finance Application / Stock Analysis Tool
- **Target Market**: South Korea
- **Primary Language**: Korean (한국어)
- **Platform**: iOS and Android mobile applications
- **Technology Stack**:
  - Frontend: React Native with Expo
  - Backend: Django REST Framework (Python) + FastAPI
  - AI: Multi-agent LLM system with OpenAI GPT-4o-mini
  - Data: PostgreSQL, Redis, ClickHouse
  - Streaming: Server-Sent Events (SSE)

## Key Features

1. **AI-Powered Analysis**: Natural language queries processed by multi-agent LLM system
2. **Real-Time Financial Data**: Live KOSPI/KOSDAQ company data from Korean Financial Supervisory Service (DART)
3. **Smart Query Routing**: Supervisor agent routes queries to specialized agents (SQL Agent, Analyzer)
4. **Streaming Responses**: Real-time token-level LLM responses via SSE
5. **Financial Metrics**: Comprehensive financial statements, ratios, and indexes
6. **RAG-Enhanced Context**: Vector search with Qdrant for company-specific insights

## Subscription Plans

- **FREE Plan**: 10 queries per day
- **PRO Plan**: 50 queries per day with priority support

## Use Cases

1. Individual investors researching Korean stocks
2. Financial analysts needing quick insights on KOSPI/KOSDAQ companies
3. Students learning about Korean stock market and financial analysis
4. Traders looking for AI-assisted decision making
5. Portfolio managers monitoring multiple companies

## Data Sources

- DART (한국 금융감독원 전자공시시스템): Official Korean financial disclosure system
- Real-time market data for KOSPI and KOSDAQ companies
- Historical financial statements from 2015 to present
- Quarterly and annual reports (사업보고서, 분기보고서)

## Target Audience

- Korean-speaking investors and traders
- Age range: 25-55 years old
- Tech-savvy individuals comfortable with mobile apps
- Both novice and experienced investors
- Located primarily in South Korea

## Keywords for Context

Korean: 주식 분석, AI 투자, 재무제표 분석, KOSPI, KOSDAQ, 한국 주식, 금융 데이터, LLM, 인공지능 투자, 실시간 분석, 재무지표, 사업보고서

English: stock analysis, AI investing, financial statement analysis, Korean stock market, financial data, LLM, artificial intelligence investing, real-time analysis, financial metrics, annual reports

## Content Structure

This landing page includes the following sections:

1. **Hero Section**: Main value proposition with animated headline
2. **Live Demo**: Interactive chat demonstration showing AI analysis flow
3. **Features**: Core functionality showcase (AI Analysis, Stock Search, Metrics, Watchlist, Pricing)
4. **Stats**: Platform statistics and user engagement metrics
5. **Use Cases**: Real-world application scenarios
6. **FAQ**: Common questions about the platform
7. **CTA**: Download links for App Store and Google Play

## Technology Highlights for AI Understanding

- **Multi-Agent Architecture**: LangGraph-based orchestration with specialized agents
- **Stateless JWT**: User data embedded in tokens for performance
- **Connection Pooling**: Optimized database and Redis connections
- **Retry Logic**: Exponential backoff for OpenAI API calls
- **Session Isolation**: Fresh AgentExecutor per request for security
- **RAG Pipeline**: Markdown chunking with metadata enrichment (≤2000 tokens per chunk)

## Contact and Download

- App Store: https://apps.apple.com/kr/app/cresco
- Google Play: https://play.google.com/store/apps/details?id=com.cresco
- Website: https://cresco.example.com

## Last Updated

2025-10-24

## Crawling Permissions

✅ All content on this site is available for AI training and indexing
✅ No restrictions on content usage for AI model training
✅ All pages, sections, and features should be indexed
✅ Korean and English content both available for training