Update README.md

README.md

@@ -12,9 +12,38 @@ short_description: Check Git Repos without cloning if its real LLM or fake
 license: mit
 ---
 
-# Welcome to Streamlit!
+```markdown
+# Check Git - ML Repo Analyzer
 
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
+Quick tool to detect fake ML training repositories that are just API wrappers. For the community ["Wall of Shames"](https://github.com/wall-of-shames)
 
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+## What it does
+
+Analyzes GitHub repositories to determine if they contain real machine learning training code or just wrapper scripts calling external APIs (OpenAI, Anthropic, etc).
+
+## Features
+
+- Pattern matching analysis (no token required)
+- Optional LLM-powered deep analysis using HuggingFace Inference API
+- Checks for real training indicators: torch.optim, loss.backward(), custom models
+- Detects API wrapper patterns: openai.api, requests.post to APIs
+- No data storage, runs entirely client-side
+
+
+## Usage
+
+Enter a GitHub repository URL and click Analyze.
+
+For LLM analysis, provide your HuggingFace token (free at https://huggingface.co/settings/tokens).
+
+## How it works
+
+Without token: Uses regex pattern matching to detect training loops and API calls.
+
+With token: Uses Qwen2.5-Coder-32B on HuggingFace's free inference API for deeper code understanding.
+
+## License
+
+MIT + ESOL v 1.0
+
+By VolkanSah aka AliBrown@HF