Bug fixes. Added missing functionality. ++

README.md

@@ -3,7 +3,7 @@
 A powerful, modern **Textual TUI** chat client with **multi-provider support** (OpenRouter, Anthropic, OpenAI, Ollama) and **MCP (Model Context Protocol)** integration, enabling AI to access local files and query SQLite databases.
 
 >[!WARNING]  
->  v3.0.0-b3 is beta software. While I strive for stability, beta versions may contain bugs, incomplete features, or unexpected behavior. I actively work on improvements and appreciate your feedback.
+>  v3.0.0-b4 is beta software. While I strive for stability, beta versions may contain bugs, incomplete features, or unexpected behavior. I actively work on improvements and appreciate your feedback.
 >
 >Beta releases are ideal for testing new features and providing feedback. For production use or maximum stability, consider using the latest stable release. 
 
@@ -146,6 +146,33 @@ On first run, you'll be prompted for your API key. Configure additional provider
 Ctrl+Q              # Quit
 ```
 
+### File Attachments
+
+Attach files to your messages using `@<file>` or `@file` syntax:
+
+```bash
+# Single file
+Describe this image @<photo.jpg>
+Analyze @~/Documents/report.pdf
+
+# Multiple files
+Compare @image1.png and @image2.png
+
+# With paths
+Review this code @./src/main.py
+What's in @/Users/username/Downloads/screenshot.png
+```
+
+**Supported file types:**
+- **Images**: `.png`, `.jpg`, `.jpeg`, `.gif`, `.webp`, `.bmp` (for vision models)
+- **PDFs**: `.pdf` (for document-capable models)
+- **Code/Text**: `.py`, `.js`, `.md`, `.txt`, `.json`, etc. (all models)
+
+**Notes:**
+- Files are automatically base64-encoded for images/PDFs
+- Maximum file size: 10MB per file
+- Works with vision-capable models (Claude, GPT-4o, Gemini, etc.)
+
 ## Web Search
 
 oAI provides universal web search capabilities for all AI providers with three options: