prism-insight

<div style="display: flex; justify-content: center; align-items: center; flex-direction: column;"> <img src="docs/images/prism-insight-logo.jpeg" alt="PRISM-INSIGHT Logo" width="300" style="margin-bottom: 20px;"> <div style="text-align: center;"> <img src="https://img.shields.io/badge/License-AGPL%20v3-blue.svg" alt="License"> <img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="Python"> <img src="https://img.shields.io/badge/OpenAI-GPT--4.1-green.svg" alt="OpenAI"> <img src="https://img.shields.io/badge/OpenAI-GPT--5-green.svg" alt="OpenAI"> <img src="https://img.shields.io/badge/Anthropic-Claude--Sonnet--4.5-green.svg" alt="Anthropic"> </div> </div> # 🔍 PRISM-INSIGHT [](https://github.com/sponsors/dragon1086) [](https://github.com/dragon1086/prism-insight/stargazers) > 📖 [한국어 문서](README_ko.md) --- ### 📊 AI-Powered Korean Stock Market Analysis & Trading System * **[Official Telegram Channel](https://t.me/prism_insight_global_en)** Features include surge-stock detection, stock analysis report downloads, trading simulations, and automated trading reports. * **➡️ Global (English) Channel:** [https://t.me/prism_insight_global_en](https://t.me/prism_insight_global_en) * **➡️ Korean Channel:** [https://t.me/stock_ai_agent](https://t.me/stock_ai_agent) * **[Official Dashboard](https://analysis.stocksimulation.kr/)**: Real-time performance dashboard for PRISM-INSIGHT live trading & simulations (Includes AI portfolio analysis, trading history, and watchlists) * **Community**: Via **[GitHub Discussions](https://github.com/dragon1086/prism-insight/discussions)** or the Telegram discussion group (@prism_insight_discuss) --- ## 📖 Project Overview PRISM-INSIGHT is a **completely open-source free project** specializing in **Korean stock market (KOSPI/KOSDAQ) analysis** through **comprehensive AI analysis agents**. It automatically detects surging Korean stocks daily through a Telegram channel, generates expert-level analyst reports, and performs trading simulations and automated trading. **✨ All features are provided 100% free!** ## 📈 Trading Simulator and Real Account Performance as of '25.11.17 ### ⭐ Season 1 (Ended '25.09.28. No real account trading) **Simulator Performance** - Start Date: 2025.03.15 - Total Trades: 51 - Profitable Trades: 23 - Loss Trades: 28 - Win Rate: 45.1% - **Cumulative Return: 408.60%** - **[Trading Performance Summary Dashboard](https://claude.ai/public/artifacts/d546cc2e-9d2c-4787-8415-86930494e198)** ### ⭐⭐ Season 2 (In Progress) **Simulator Performance** - Start Date: 2025.09.29 - Total Trades: 16 - Profitable Trades: 10 - Loss Trades: 6 - Win Rate: 62.50% - **Total Cumulative Return from Sold Stocks: 110.99%** - **Realized Portfolio Return: 11.10%** (managed across 10 slots, 110.99% ÷ 10) - Market Benchmark (from Season 2 start): KOSPI +16.91%, KOSDAQ +6.05% - **[Trading Performance Summary Dashboard](https://analysis.stocksimulation.kr/)** **Real Account Performance** - Start Date: 2025.09.29 - Initial Capital: ₩9,969,801 - Current Total Assets (Valuation + Cash): ₩10,901,561 - **Return: +9.35%** ## 🤖 AI Agent System Architecture (Core Feature) PRISM-INSIGHT is a **multi-agent system where 13 specialized AI agents collaborate**. Each agent specializes in a specific analysis domain and works organically together to deliver expert-level comprehensive analysis and trading. ### 📊 Analysis Team (6 Agents) - GPT-4.1 Based #### 1. Technical Analyst <img src="docs/images/aiagent/technical_analyst.jpeg" alt="Technical Analyst" width="300"/> - **Role**: Stock price and trading volume technical analysis expert - **Analysis Items**: - Price trends, moving averages, support/resistance levels - Chart patterns and technical indicators (RSI, MACD, Bollinger Bands) - Technical perspective #### 2. Trading Flow Analyst <img src="docs/images/aiagent/tranding_flow_analyst.jpeg" alt="Trading Flow Analyst" width="300"/> - **Role**: Investor trading trend analysis expert - **Analysis Items**: - Trading patterns of institutional/foreign/individual investors - Investor group trends through volume analysis #### 3. Financial Analyst <img src="docs/images/aiagent/financial_analyst.jpeg" alt="Financial Analyst" width="300"/> - **Role**: Corporate finance and valuation analysis expert - **Analysis Items**: - Financial statement analysis (revenue, operating profit, net income) - Valuation assessment (PER, PBR, ROE, etc.) - Target price and securities firm consensus #### 4. Industry Analyst <img src="docs/images/aiagent/industry_analyst.jpeg" alt="Industry Analyst" width="300"/> - **Role**: Corporate business structure and competitiveness analysis expert - **Analysis Items**: - Business portfolio and market share - Strengths/weaknesses compared to competitors - R&D investment and growth drivers #### 5. Information Analyst <img src="docs/images/aiagent/information_analyst.jpeg" alt="Information Analyst" width="300"/> - **Role**: News and issue trend analysis expert - **Analysis Items**: - Identifying causes of same-day stock price fluctuations - Latest news and disclosure analysis - Industry trends and political/economic issues #### 6. Market Analyst <img src="docs/images/aiagent/market_analyst.jpeg" alt="Market Analyst" width="300"/> - **Role**: Overall market and macroeconomic analysis expert - **Analysis Items**: - KOSPI/KOSDAQ index analysis - Macroeconomic indicators (interest rates, exchange rates, prices) - Correlation between global economy and Korean market --- ### 💡 Strategy Team (1 Agent) - GPT-4.1 Based #### 7. Investment Strategist <img src="docs/images/aiagent/investment_strategist.jpeg" alt="Investment Strategist" width="300"/> - **Role**: Integrates all analysis results to establish final investment strategy - **Provides**: - Customized strategies for short/medium/long-term investors - Risk level and trading timing suggestions - Comprehensive opinion from portfolio perspective --- ### 💬 Communication Team (3 Agents) - GPT-4.1 / GPT-5-nano #### 8-1. Summary Specialist <img src="docs/images/aiagent/summary_specialist.jpeg" alt="Summary Specialist" width="300"/> - **Role**: Converts detailed reports into core summaries for investors - **Features**: - Generates concise Telegram messages within 400 characters - Extracts key information and investment points - Telegram-optimized formatting #### 8-2. Quality Inspector <img src="docs/images/aiagent/quality_inspector.jpeg" alt="Quality Inspector" width="300"/> - **Role**: Evaluates quality of generated messages and suggests improvements - **Features**: - Verifies accuracy, clarity, and format compliance - Detects hallucinations and identifies errors - Collaborates with Summary Specialist for iterative improvement to EXCELLENT rating #### 8-3. Translation Specialist <img src="docs/images/aiagent/translator_specialist.png" alt="Translation Specialist" width="300"/> - **Role**: Translates analysis reports and messages to multiple languages - **Features**: - Supports multi-language broadcasting (English, Japanese, Chinese, etc.) - Preserves technical terminology and market context - Enables parallel transmission to language-specific Telegram channels --- ### 📈 Trading Simulation Team (2 Agents) - GPT-5 Based #### 9-1. Buy Specialist <img src="docs/images/aiagent/buy_specialist.jpeg" alt="Buy Specialist" width="300"/> - **Role**: Buy decision-making and entry management based on AI reports - **Features**: - Evaluates buy score based on valuation and momentum (1-10 points) - Manages portfolio with maximum 10 slots - Industry diversification and risk management - Dynamic target/stop-loss setting - Detailed trading scenario creation #### 9-2. Sell Specialist <img src="docs/images/aiagent/sell_specialist.jpeg" alt="Sell Specialist" width="300"/> - **Role**: Monitors holdings based on trading scenarios and determines sell timing - **Features**: - Real-time monitoring of stop-loss/profit-taking scenarios - Technical trend and market environment analysis - Portfolio optimization adjustment suggestions - Prudent decisions considering 100% exit characteristics --- ### 💬 User Consultation Team (2 Agents) - Claude Sonnet 4.5 Based #### 10-1. Portfolio Consultant <img src="docs/images/aiagent/portfolio_consultant.jpeg" alt="Portfolio Consultant" width="300"/> - **Role**: User portfolio evaluation and customized investment advice - **Features**: - Analysis based on user's average purchase price and holding period - Comprehensive evaluation using latest market data and news - Adaptive responses to user request styles (friendly/expert/direct, etc.) - Customized advice for profit/loss positions #### 10-2. Dialogue Manager <img src="docs/images/aiagent/dialogue_manager.jpeg" alt="Dialogue Manager" width="300"/> - **Role**: Maintains conversation context and handles follow-up questions - **Features**: - Remembers and references previous conversation context - Consistent answers to additional questions - Additional data lookup when necessary - Maintains natural conversation flow --- ## 🔄 Agent Collaboration Workflow <img src="docs/images/aiagent/agent_workflow2.png" alt="Agent Workflow" width="700"> ## 🎯 Key Features - **🤖 AI Comprehensive Analysis (Core)**: Expert-level stock analysis through GPT-4.1 based multi-agent system [](https://youtu.be/4WNtaaZug74) - **📊 Automatic Surge Stock Detection**: Watchlist selection through hourly (morning/afternoon) market trend analysis <img src="docs/images/trigger-en.png" alt="Surge Stock Detection" width="500"> - **📱 Automatic Telegram Transmission**: Real-time transmission of analysis results to Telegram channel <img src="docs/images/summary-en.png" alt="Summary Transmission" width="500"> - **📈 Trading Simulation**: Investment strategy simulation using GPT-5 based generated reports <img src="docs/images/simulation1-en.png" alt="Simulation 1" width="500"> <img src="docs/images/simulation2-en.png" alt="Simulation 2" width="500"> <img src="docs/images/season1_dashboard.png" alt="Simulation Performance" width="500"> - **💱 Automated Trading**: Automatic trading according to trading simulation results through Korea Investment & Securities API - **🎨 Realtime Dashboard**: We transparently disclose all information on the AI-traded portfolio, its performance relative to the market, the AI's trading rationale, full trading history, watchlist, and system maintenance costs. <img src="docs/images/dashboard1-en.png" alt="Dashboard 1" width="500"> <img src="docs/images/dashboard2-en.png" alt="Dashboard 2" width="500"> <img src="docs/images/dashboard3-en.png" alt="Dashboard 3" width="500"> <img src="docs/images/dashboard4-en.png" alt="Dashboard 4" width="500"> <img src="docs/images/dashboard5-en.png" alt="Dashboard 5" width="500"> <img src="docs/images/dashboard6-en.png" alt="Dashboard 6" width="500"> <img src="docs/images/dashboard7-en.png" alt="Dashboard 7" width="500"> - **🎬 YouTube Event Fund Crawler** (NEW): Contrarian investment strategy based on '전인구경제연구소' YouTube channel analysis - Automatic monitoring of new videos via RSS feed - Audio extraction and transcription using OpenAI Whisper API - AI-powered content analysis for market sentiment detection - Contrarian investment recommendations (inverse/leveraged ETF suggestions) - 📖 See [YOUTUBE_EVENT_FUND_CRAWLER.md](events/YOUTUBE_EVENT_FUND_CRAWLER.md) for details ## 🧠 AI Model Usage - **Core Analysis**: OpenAI GPT-4.1 (Comprehensive stock analysis agent) - **Trading Simulation**: OpenAI GPT-5 (Investment strategy simulation) - **Telegram Conversation**: Anthropic Claude Sonnet 4.5 (Bot interaction) - **Translation**: OpenAI GPT-5-NANO (Multilingual broadcasting on a Telegram channel) ## 💡 MCP Servers Used - **[kospi_kosdaq](https://github.com/dragon1086/kospi-kosdaq-stock-server)**: MCP server for KRX (Korea Exchange) stock data in report generation - **[firecrawl](https://github.com/mendableai/firecrawl-mcp-server)**: Web crawling specialized MCP server for report generation - **[perplexity](https://github.com/perplexityai/modelcontextprotocol/tree/main)**: Web search specialized MCP server for report generation - **[sqlite](https://github.com/modelcontextprotocol/servers-archived/tree/HEAD/src/sqlite)**: MCP server specialized in internal DB storage for trading simulation records - **[time](https://github.com/modelcontextprotocol/servers/tree/main/src/time)**: MCP server for current time retrieval ## 🚀 Getting Started ### Prerequisites - Python 3.10+ - OpenAI API Key (GPT-4.1, GPT-5) - Anthropic API Key (Claude-Sonnet-4.5) - Telegram Bot Token and Channel ID - Playwright (for PDF conversion) - Korea Investment & Securities API app key and secret key ### Installation 1. **Clone Repository** ```bash git clone https://github.com/dragon1086/prism-insight.git cd prism-insight ``` 2. **Install Dependencies** ```bash pip install -r requirements.txt ``` 3. **Prepare Configuration Files** Copy the example files to create actual configuration files: ```bash cp .env.example .env cp ./examples/streamlit/config.py.example ./examples/streamlit/config.py cp mcp_agent.config.yaml.example mcp_agent.config.yaml cp mcp_agent.secrets.yaml.example mcp_agent.secrets.yaml ``` 4. **Edit Configuration Files** Edit the copied configuration files to enter necessary API keys and settings. 5. **Install Playwright** (for PDF conversion) The system will **automatically install** Playwright browser on first run. For manual installation: ```bash # Install Playwright package (included in requirements.txt) pip install playwright # Download Chromium browser python3 -m playwright install chromium ``` **Platform-specific installation:** ```bash # macOS pip3 install playwright python3 -m playwright install chromium # Ubuntu/Debian pip install playwright python3 -m playwright install --with-deps chromium # Rocky Linux 8 / CentOS / RHEL pip3 install playwright python3 -m playwright install --with-deps chromium # Or use the installation script cd utils chmod +x setup_playwright.sh ./setup_playwright.sh ``` **📖 For detailed installation instructions, see:** [utils/PLAYWRIGHT_SETUP.md](utils/PLAYWRIGHT_SETUP.md) 6. **Install perplexity-ask MCP Server** ```bash cd perplexity-ask npm install ``` 7. **Install Korean Fonts** (Linux environment) Korean fonts are required for Korean text display in charts on Linux. ```bash # Rocky Linux 8 / CentOS / RHEL sudo dnf install google-nanum-fonts # Ubuntu 22.04+ / Debian Run ./cores/ubuntu_font_installer.py # Refresh font cache sudo fc-cache -fv python3 -c "import matplotlib.font_manager as fm; fm.fontManager.rebuild()" Note: macOS and Windows have default Korean font support, no installation needed ``` 8. **Auto-run Setup (Crontab)** Set up crontab to run automatically: ```bash # Simple setup (recommended) chmod +x utils/setup_crontab_simple.sh utils/setup_crontab_simple.sh # Or advanced setup chmod +x utils/setup_crontab.sh utils/setup_crontab.sh ``` See [CRONTAB_SETUP.md](utils/CRONTAB_SETUP.md) for details. ### Required Configuration Files The following configuration files must be set up to run the project: #### 🔧 Core Settings (Required) - **`mcp_agent.config.yaml`**: MCP agent configuration - **`mcp_agent.secrets.yaml`**: MCP agent secret information (API keys, etc.) #### 📱 Telegram Settings (Optional) - **`.env`**: Environment variables including Telegram channel ID, bot token, etc. - Use `--no-telegram` option to run without Telegram - All analysis features work normally without Telegram #### 🌐 Web Interface Settings (Optional) - **`./examples/streamlit/config.py`**: Report generation web settings 💡 **Tip**: Use `--no-telegram` option to run without `.env` file! ## 📋 Usage ### Basic Execution Run the entire pipeline to automate from surge stock analysis to Telegram transmission: ```bash # Run both morning + afternoon (Telegram enabled) python stock_analysis_orchestrator.py --mode both # Morning only python stock_analysis_orchestrator.py --mode morning # Afternoon only python stock_analysis_orchestrator.py --mode afternoon # Local test without Telegram (no Telegram setup needed) python stock_analysis_orchestrator.py --mode morning --no-telegram # Generate English reports (default: Korean) python stock_analysis_orchestrator.py --mode morning --language en # Broadcast to multiple language channels (requires setup in .env) python stock_analysis_orchestrator.py --mode morning --broadcast-languages en,ja,zh ``` #### 💡 Telegram Option (`--no-telegram`) You can run the system without Telegram setup: **Usage Scenarios:** - 🧪 **Local Development/Testing**: Quickly test core features without Telegram setup - 🚀 **Performance Optimization**: Skip message generation and transmission process - 🔧 **Debugging**: Focus only on analysis and report generation features **Execution Effects:** - ✅ Surge stock detection → Report generation → PDF conversion → Tracking system (all working normally) - ❌ Telegram alerts, message generation, message transmission (skipped) - 💰 AI summary generation cost savings **Required Environment Variables (when using Telegram):** ```bash # .env file TELEGRAM_CHANNEL_ID="-1001234567890" # Main channel (Korean by default) TELEGRAM_BOT_TOKEN="1234567890:ABCdefGHIjklMNOpqrsTUVwxyz" # Multi-language broadcasting (optional) # Use with --broadcast-languages argument (e.g., --broadcast-languages en,ja,zh) # TELEGRAM_CHANNEL_ID_EN="-1001234567891" # English channel # TELEGRAM_CHANNEL_ID_JA="-1001234567892" # Japanese channel # TELEGRAM_CHANNEL_ID_ZH="-1001234567893" # Chinese channel ``` ### Individual Module Execution **1. Run Surge Stock Detection Only** ```bash python trigger_batch.py morning INFO --output trigger_results.json ``` **2. Generate AI Analysis Report for Specific Stock (Core Feature)** ```bash python cores/main.py # Or use analyze_stock function directly ``` **3. PDF Conversion** ```bash python pdf_converter.py input.md output.pdf ``` **4. Generate and Send Telegram Messages** ```bash python telegram_summary_agent.py python telegram_bot_agent.py ``` ## 📁 Project Structure ``` prism-insight/ ├── 📂 cores/ # 🤖 Core AI Analysis Engine │ ├── 📂 agents/ # AI Agent Modules │ │ ├── company_info_agents.py # Company Information Analysis Agent │ │ ├── news_strategy_agents.py # News and Investment Strategy Agent │ │ ├── stock_price_agents.py # Stock Price and Volume Analysis Agent │ │ ├── telegram_quality_inspector.py # Quality Inspector Agent │ │ ├── telegram_summary_agent.py # Summary Specialist Agent │ │ └── telegram_translator_agent.py # Translation Specialist Agent │ ├── analysis.py # Comprehensive Stock Analysis (Core) │ ├── main.py # Main Analysis Execution │ ├── report_generation.py # Report Generation │ ├── stock_chart.py # Chart Generation │ └── utils.py # Utility Functions ├── 📂 examples/streamlit/ # Web Interface ├── 📂 trading/ # 💱 Automated Trading System (Korea Investment & Securities API) │ ├── kis_auth.py # KIS API Authentication and Token Management │ ├── domestic_stock_trading.py # Domestic Stock Trading Core Module │ ├── portfolio_telegram_reporter.py # Portfolio Telegram Reporter │ ├── 📂 config/ # Configuration File Directory │ │ ├── kis_devlp.yaml # KIS API Configuration (app key, account number, etc.) │ │ └── kis_devlp.yaml.example # Configuration File Example │ └── 📂 samples/ # API Sample Code ├── 📂 utils/ # Utility Scripts ├── 📂 tests/ # Test Code ├── stock_analysis_orchestrator.py # 🎯 Main Orchestrator ├── telegram_config.py # Telegram Configuration Management Class ├── trigger_batch.py # Surge Stock Detection Batch ├── telegram_bot_agent.py # Telegram Bot (Claude Based) ├── stock_tracking_agent.py # Trading Simulation (GPT-5) ├── stock_tracking_enhanced_agent.py # Enhanced Trading Simulation ├── pdf_converter.py # PDF Conversion ├── requirements.txt # Dependency List ├── .env.example # Environment Variable Example ├── mcp_agent.config.yaml.example # MCP Agent Configuration Example ├── mcp_agent.secrets.yaml.example # MCP Agent Secret Example ``` ## 📈 Analysis Report Composition The comprehensive analyst report generated by AI agents consists of the following sections: 1. **📊 Key Investment Points** - Summary and main points 2. **📈 Technical Analysis** - Stock price and trading volume analysis - Investor trading trend analysis 3. **🏢 Fundamental Analysis** - Company status analysis - Company overview analysis 4. **📰 News Trend Analysis** - Recent major news and issues 5. **🌐 Market Analysis** - KOSPI/KOSDAQ index and macro-environment analysis 6. **💡 Investment Strategy and Opinion** - Strategies for different investor types ## 🔧 Customization ### Modify Surge Stock Detection Criteria In `trigger_batch.py`, you can modify: - Trading volume increase threshold - Stock price increase criteria - Market capitalization filtering conditions ### Modify AI Prompts You can customize analysis instructions in each agent file in the `cores/agents/` directory. ### Change Chart Style You can modify chart colors, styles, and indicators in `cores/stock_chart.py`. ## 🤝 Contributing 1. Fork the project 2. Create a feature branch (`git checkout -b feature/amazing-feature`) 3. Commit your changes (`git commit -m 'Add amazing feature'`) 4. Push to the branch (`git push origin feature/amazing-feature`) 5. Create a Pull Request ## 📄 License **PRISM-INSIGHT is dual-licensed:** ### For Individual & Open-Source Use [](https://www.gnu.org/licenses/agpl-3.0) This project is licensed under the **GNU Affero General Public License v3.0 (AGPL-3.0)** for: - Individual personal use - Non-commercial projects - Open-source development - Academic research The AGPL-3.0 license requires: - Source code disclosure for any modifications - Source code availability for network/service use - Same license for derived works 📖 **[Full AGPL-3.0 License Text](LICENSE)** ### For Commercial SaaS Use For SaaS companies and commercial use, a separate **Commercial License** is required. **Commercial License Benefits:** - No obligation to disclose source code - Freedom to modify without publishing changes - Proprietary integration allowed - Dedicated technical support - Priority feature development **Pricing:** - Startups (<50 employees): $500/month - SME (50-500 employees): $2,000/month - Enterprise (>500 employees): Custom pricing - **Exclusive SaaS License:** Starting from $500/month (1-year commitment) 📧 **Contact for Commercial Licensing:** dragon1086@naver.com 📄 **[Commercial License Details](LICENSE-COMMERCIAL.md)** ### Why Dual License? **For the Community:** - Free for personal learning and development - Contribute to open-source ecosystem - Full transparency with AGPL-3.0 **For Sustainability:** - Commercial licenses fund ongoing development - Maintain server infrastructure ($174/month operating costs) - Add new features and AI models - Provide quality support **For Businesses:** - Legal clarity for commercial use - No source code disclosure requirements - Professional support and SLA - Customization opportunities --- **Questions about licensing?** - 📧 Email: dragon1086@naver.com - 💬 Telegram: [@stock_ai_agent](https://t.me/stock_ai_agent) - 🐛 GitHub Issues: [Report here](https://github.com/dragon1086/prism-insight/issues) ## ⚠️ Disclaimer The analysis information provided by this system is for investment reference purposes only and is not intended as investment advice. All investment decisions and resulting profits/losses are the responsibility of the investor. ## 📞 Contact For project inquiries or bug reports, please submit through [GitHub Issues](https://github.com/dragon1086/prism-insight/issues). ## 💝 Sponsorship for Project Sustainability ### 🥇 Seeking Gold Sponsor (1 spot only, now recruiting) **Help PRISM-INSIGHT run permanently for $500/month.** ✨ **Gold Sponsor Exclusive Benefits:** - 🏆 **Exclusive Exposure**: Logo placed exclusively at the top of GitHub README + [Dashboard](https://analysis.stocksimulation.kr/) main - 📊 **User Reach**: Continuous brand exposure to 450+ Telegram subscribers + daily growing GitHub visitors - 💎 **Honor**: The only official sponsorship partner of a completely open-source AI stock analysis system - 🛡️ **Stability Guarantee**: Provide an environment where a developer short on time due to parenting can focus on system maintenance - 🤝 **Direct Communication**: Opportunity to provide input on project roadmap and priorities --- **$500/month Details** 1. AI Operation Cost Coverage: Full API cost for premium AI models like GPT-5 2. Stabilization Compensation: Compensation for emergency maintenance and stabilization work on core systems --- 📧 **Gold Sponsor Partnership Inquiry**: - **Email**: dragon1086@naver.com - **GitHub Issues**: [Partnership Inquiry](https://github.com/dragon1086/prism-insight/issues/new?labels=sponsorship&template=partnership.md) - **Telegram**: @stock_ai_ko - Or select $500 tier on [GitHub Sponsors](https://github.com/sponsors/dragon1086) > 💡 **Need a customized proposal for your company?** Contact us via email for consultation first. --- ### 💙 To Individual Sponsors Even small support gives great strength! Support the project with the cost of a cup of coffee. <div align="center"> <a href="https://github.com/sponsors/dragon1086"> <img src="https://img.shields.io/badge/Sponsor_on_GitHub-❤️-ff69b4?style=for-the-badge&logo=github-sponsors" alt="Sponsor on GitHub"> </a> </div> ### 💰 Transparent Operations Monthly costs of approximately ₩260,000 for API and server expenses (as of October '25): - OpenAI API (GPT-4.1, GPT-5): ~₩140,000/month - Anthropic API (Claude Sonnet 4.5): ~₩30,000/month - Firecrawl API (MCP Server): ~₩30,000/month - Perplexity API (MCP Server): ~₩15,000/month - Server and Infrastructure: ~₩45,000/month Currently used by 450 people for free. ### ✨ Current Sponsorship Status Thank you so much! PRISM-INSIGHT continues to operate with your support. #### 🥇 Gold Sponsor <!-- gold-sponsor --> **Spot available!** Be our first partner. <!-- gold-sponsor --> #### 💙 Individual Sponsors <!-- sponsors --> **Core Supporter** - [@tehryung-ray](https://github.com/tehryung-ray) 💙 **Supporter** - [@ddkerty](https://github.com/ddkerty) 💙 - [@jk5745](https://github.com/jk5745) 💙 <!-- sponsors --> --- **Important:** All features are provided free regardless of sponsorship. Sponsorship is just support for service continuity. --- ## ⭐ Project Growth Achieved **250+ Stars in just 10 weeks** since mid-August '25 launch! [](https://star-history.com/#dragon1086/prism-insight&Date) --- **⭐ If this project helped you, please give us a Star!**