Update README.md

README.md

@@ -1,316 +1,11 @@
-<div align="center">
-
-<img src="docs/assets/logo.png" alt="Wallstreet-AI Logo" />
-
-# Wallstreet-AI
-
-**An agentic financial analysis assistant powered by legendary investor personas.**
-
-Combine the investment philosophies of Warren Buffett, Charlie Munger, and other market legends with a structured AI pipeline — delivering SWOT analyses, technical reports, earnings breakdowns, and more, all streamed in real time.
-
-[![Python](https://img.shields.io/badge/Python-3.10%2B-3776AB?style=flat-square&logo=python&logoColor=white)](https://www.python.org/)
-[![FastAPI](https://img.shields.io/badge/FastAPI-SSE%20Streaming-009688?style=flat-square&logo=fastapi&logoColor=white)](https://fastapi.tiangolo.com/)
-[![Gradio](https://img.shields.io/badge/Gradio-Web%20UI-FF7C00?style=flat-square&logo=gradio&logoColor=white)](https://gradio.app/)
-[![OpenAI](https://img.shields.io/badge/OpenAI-Responses%20API-412991?style=flat-square&logo=openai&logoColor=white)](https://platform.openai.com/)
-
-[![deepwiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/davidkim205/wallstreet-ai)
-
-[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1GUbw0Ef0bJQfkddrDNCuvADXG3ujxo9H)
-[![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/davidkim205/wallstreet-ai)
-
-<p align="center">
-  <a href="README.md">🇺🇸 English</a> |
-  <a href="README_KR.md">🇰🇷 한국어</a>
-</p>
-</div>
-
 ---
-
-## Why Wallstreet-AI?
-
-Traditional financial tools give you data. Wallstreet-AI gives you *interpretation*.
-
-Instead of reading raw numbers, you ask a question the way you'd ask a mentor — and receive a structured, persona-driven analysis that thinks through the data the way a legendary investor would.
-
-- Warren Buffett would ask: *"Does this business have a durable moat?"*
-- Charlie Munger would ask: *"Am I avoiding the big mistakes?"*
-- You just ask: *"Should I be worried about Apple's margins?"*
-
----
-
-## What is Wallstreet-AI?
-
-Wallstreet-AI turns a plain-language question like *"What would Warren Buffett think of Apple's latest earnings?"* into a fully structured analysis pipeline:
-
-1. **Intent parsing** — extracts ticker, analysis type, time period, and earnings quarter from your question
-2. **Tool routing** — selects only the data sources required for that analysis type
-3. **Data collection** — fetches prices, fundamentals, technicals, earnings, and news via `yfinance` and Google News RSS
-4. **News enrichment** — scrapes article bodies to build real-time news context
-5. **LLM generation** — streams a tailored report through the OpenAI Responses API
-6. **Result storage** — saves every analysis and persona to JSONL for later review
-
-You can apply an investor persona at any stage. The same underlying data produces a completely different report depending on whether you view it through a value-investing lens or a growth-focused one.
-
----
-
-## Demo
-
-### Streaming stock analysis
-
-![Streaming stock analysis with pipeline progress and final results](docs/assets/analyze.gif)
-
-### Persona generation
-
-![Persona generation and saving based on a financial figure](docs/assets/persona.gif))
-
----
-
-## Key Features
-
-| Feature | Details |
-|---|---|
-| **Natural-language intent parsing** | Extracts ticker, analysis type, period, and year/quarter automatically |
-| **7 analysis report types** | General · Technical · Fundamental · Earnings · SWOT · News summary · Comparative |
-| **Investor personas** | Warren Buffett, Charlie Munger, and any figure you define — tone and reasoning style adapt to the persona |
-| **Technical indicators** | RSI, MACD, moving averages (SMA/EMA), Bollinger Bands |
-| **Real-time news context** | Google News RSS + `trafilatura` article body extraction |
-| **SSE streaming** | Token-by-token response streaming via FastAPI Server-Sent Events |
-| **Three entry points** | CLI · FastAPI REST API · Gradio web UI |
-| **JSONL logging** | All analyses and personas persisted for reproducibility |
-
----
-
-## Quick Start
-
-> Requires Python 3.10+ and an OpenAI API key.
-
-```bash
-# 1. Clone and enter the repo
-git clone https://github.com/davidkim205/wallstreet-ai-ai.git
-cd wallstreet-ai
-
-# 2. Create a virtual environment
-uv venv
-source .venv/bin/activate   # Windows: .venv\Scripts\activate
-
-# 3. Install dependencies
-uv pip install -r requirements.txt
-
-# 4. Set up environment variables
-cp .env.example .env
-# Open .env and add your LLM_MODEL_API_KEY
-
-# 5a. Run the FastAPI server (for web UI or API access)
-uvicorn api_server:app --reload
-
-# 5b. In a second terminal, launch the Gradio UI
-python gradio_app.py --api-url http://127.0.0.1:8000/analyze/
-```
-
-Then open your browser at `http://localhost:7860` and ask anything:
-
-> *"Give me a Warren Buffett-style fundamental analysis of Microsoft for the last 2 years"*
-
-To use the CLI only, skip step 5a and run `python pipeline.py` directly.
-
----
-## Online Demo
-
-You can try Wallstreet-AI instantly without installing anything.<br>
-Click the platform name to open.
-
-| Platform | Description |
-|---|---|
-| [**Google Colab**](https://colab.research.google.com/drive/1GUbw0Ef0bJQfkddrDNCuvADXG3ujxo9H) | Run the full pipeline in a hosted notebook.<br> Installs the GitHub repo and allows interactive testing. |
-| [**HuggingFace Spaces**](https://huggingface.co/spaces/davidkim205/wallstreet-ai) | Live web demo similar to Gradio UI for quick experimentation. |
-
----
-
-## Installation
-
-### Python environment
-
-Python 3.10 or later is recommended. Using `uv` is strongly advised for faster installs:
-
-```bash
-uv venv
-source .venv/bin/activate
-uv pip install -r requirements.txt
-```
-
-### Environment variables
-
-```bash
-cp .env.example .env
-```
-
-Edit `.env` with your credentials:
-
-```env
-LLM_MODEL_NAME=gpt-4o-mini
-LLM_MODEL_API_KEY=<your OpenAI API key>
-LOG_FILE=analysis_results.jsonl
-PERSONA_FILE=persona.jsonl
-```
-
-| Variable | Description |
-|---|---|
-| `LLM_MODEL_NAME` | Model name passed to the OpenAI client |
-| `LLM_MODEL_API_KEY` | OpenAI API key (falls back to `OPENAI_API_KEY` if unset) |
-| `LOG_FILE` | Path for analysis result logs (JSONL) |
-| `PERSONA_FILE` | Path for saved investor personas (JSONL) |
-
----
-
-## How to Run
-
-### CLI
-
-```bash
-python pipeline.py
-```
-
-At startup the CLI lists saved personas. Enter a persona number to apply it, or press Enter to skip. After entering your question the pipeline runs and prints the full report plus a data summary.
-
-Exit with `exit`, `quit`, or `종료`.
-
-### FastAPI server
-
-```bash
-uvicorn api_server:app --reload
-```
-
-Starts at `http://127.0.0.1:8000`. Available endpoints:
-
-| Method | Endpoint | Description |
-|---|---|---|
-| `POST` | `/analyze/` | Run a full analysis (streaming or batch) |
-| `POST` | `/persona/` | Generate and save a new investor persona |
-
-### Gradio UI
-
-Start the FastAPI server first, then:
-
-```bash
-python gradio_app.py --api-url http://127.0.0.1:8000/analyze/
-```
-
-Default binding: `0.0.0.0:7860`
-
-| Option | Description |
-|---|---|
-| `--api-url` | SSE endpoint for the analysis pipeline |
-| `--port` | Gradio server port |
-| `--server-name` | Binding address |
-| `--share` | Generate a public Gradio share link |
-
-The UI has three tabs: **Ask a Question**, **Create a Persona**, and **Investor Profile**.
-
----
-
-## API Reference
-
-### Analysis — batch mode
-
-```bash
-curl -X POST "http://127.0.0.1:8000/analyze/" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "query": "Summarize AAPL recent earnings and key investment points",
-    "stream": false
-  }'
-```
-
-Response:
-
-```json
-{
-  "type": "result",
-  "query": "Summarize AAPL recent earnings and key investment points",
-  "ticker": "AAPL",
-  "analysis_type": "earnings",
-  "data_context": {},
-  "llm_response": "...",
-  "timestamp": "2026-03-30 10:00:00",
-  "stdout": "..."
-}
-```
-
-### Analysis — streaming mode
-
-```bash
-curl -N -X POST "http://127.0.0.1:8000/analyze/" \
-  -H "Content-Type: application/json" \
-  -d '{"query": "Summarize AAPL recent earnings", "stream": true}'
-```
-
-SSE event sequence:
-
-```
-data: {"type":"status","message":"Parsing intent..."}
-data: {"type":"stdout","message":"[③] Collecting data (ticker=AAPL, period=1y)..."}
-data: {"type":"delta","delta":"Apple's most recent quarter shows margin expansion..."}
-data: {"type":"result","ticker":"AAPL","analysis_type":"earnings", ...}
-data: {"type":"done"}
-```
-
-| Event type | Meaning |
-|---|---|
-| `status` | High-level pipeline progress |
-| `stdout` | Server-side log lines |
-| `delta` | Incremental model output tokens |
-| `result` | Final structured payload |
-| `done` | Stream complete |
-
-### Analysis with a persona
-
-```bash
-curl -X POST "http://127.0.0.1:8000/analyze/" \
-  -H "Content-Type: application/json" \
-  -d '{
-    "query": "SWOT analysis of Samsung Electronics",
-    "stream": false,
-    "persona_name": "Warren Buffett"
-  }'
-```
-
-### Create a persona
-
-```bash
-curl -X POST "http://127.0.0.1:8000/persona/" \
-  -H "Content-Type: application/json" \
-  -d '{"info": "Warren Buffett"}'
-```
-
----
-
-## Persona System
-
-The persona system rewrites the analysis prompt to reflect a specific investor's voice and reasoning style. When a persona is active, the model receives:
-
-- A summary of the figure's background and track record
-- Their core financial mindset and mental models
-- Their preferred data analysis style (e.g. DCF-focused vs. narrative-focused)
-- Their typical response tone and vocabulary
-- Core investment principles
-- Representative quotes
-
----
-
-## Contributing
-
-Contributions are welcome. Please open an issue first to discuss significant changes. For small fixes, a pull request with a clear description is sufficient.
-
----
-
-## License
-
-See [LICENSE](LICENSE) for details.
-
----
-
-<div align="center">
-
-*Ask any stock question. Get an answer that thinks like the greats.*
-
-</div>
+title: Wallstreet Ai Test
+emoji: 📉
+colorFrom: green
+colorTo: pink
+sdk: gradio
+sdk_version: "4.44.1"
+app_file: app.py
+pinned: false
+short_description: test ask q
+---