diff --git a/README.md b/README.md
index 2b397e6..cb4add3 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,8 @@
 # EmbeddingBuddy
 
-A Python Dash application for interactive exploration and visualization of
-embedding vectors through dimensionality reduction techniques.
+A web application for interactive exploration and visualization of embedding
+vectors through dimensionality reduction techniques. Compare documents and prompts
+in the same embedding space to understand semantic relationships.
 
 ![Screenshot of 3d graph and UI for Embedding Buddy](./embedding-buddy-screenshot.png)
 
@@ -9,28 +10,45 @@ embedding vectors through dimensionality reduction techniques.
 
 EmbeddingBuddy provides an intuitive web interface for analyzing high-dimensional
 embedding vectors by applying various dimensionality reduction algorithms and
-visualizing the results in interactive 2D and 3D plots.
+visualizing the results in interactive 2D and 3D plots. The application supports
+dual dataset visualization, allowing you to compare documents and prompts to
+understand how queries relate to your content.
 
 ## Features
 
-- **Dimensionality Reduction**: Support for PCA, t-SNE, and UMAP algorithms
-- **Interactive Visualizations**: 2D and 3D plots using Plotly
-- **Web Interface**: Built with Python Dash for easy accessibility
-- **Vector Analysis**: Tools for exploring embedding vector relationships and
-  patterns
+- **Dual file upload** - separate drag-and-drop for documents and prompts
+- **Multiple dimensionality reduction methods**: PCA, t-SNE, and UMAP
+- **Interactive 2D/3D visualizations** with toggle between views
+- **Color coding options** by category, subcategory, or tags
+- **Visual distinction**: Documents appear as circles, prompts as diamonds with desaturated colors
+- **Prompt visibility toggle** - show/hide prompts to reduce visual clutter
+- **Point inspection** - click points to view full content and identify document vs prompt
+- **Reset functionality** - clear all data to start fresh
+- **Sidebar layout** with controls on left, large visualization area on right
+- **Real-time visualization** optimized for small to medium datasets
 
 ## Data Format
 
-EmbeddingBuddy accepts newline-delimited JSON (NDJSON) files where each line contains an embedding document with the following structure:
+EmbeddingBuddy accepts newline-delimited JSON (NDJSON) files for both documents
+and prompts. Each line contains an embedding with the following structure:
+
+**Documents:**
 
 ```json
 {"id": "doc_001", "embedding": [0.1, -0.3, 0.7, ...], "text": "Sample text content", "category": "news", "subcategory": "politics", "tags": ["election", "politics"]}
 {"id": "doc_002", "embedding": [0.2, -0.1, 0.9, ...], "text": "Another example", "category": "review", "subcategory": "product", "tags": ["tech", "gadget"]}
 ```
 
+**Prompts:**
+
+```json
+{"id": "prompt_001", "embedding": [0.15, -0.28, 0.65, ...], "text": "Find articles about machine learning applications", "category": "search", "subcategory": "technology", "tags": ["AI", "research"]}
+{"id": "prompt_002", "embedding": [0.72, 0.18, -0.35, ...], "text": "Show me product reviews for smartphones", "category": "search", "subcategory": "product", "tags": ["mobile", "reviews"]}
+```
+
 **Required Fields:**
 
-- `embedding`: Array of floating-point numbers representing the vector
+- `embedding`: Array of floating-point numbers representing the vector (must be same dimensionality for both documents and prompts)
 - `text`: String content associated with the embedding
 
 **Optional Fields:**
@@ -40,15 +58,7 @@ EmbeddingBuddy accepts newline-delimited JSON (NDJSON) files where each line con
 - `subcategory`: Secondary classification
 - `tags`: Array of string tags for flexible labeling
 
-## Features
-
-- **Drag-and-drop file upload** for NDJSON embedding datasets
-- **Multiple dimensionality reduction methods**: PCA, t-SNE, and UMAP
-- **Interactive 2D/3D visualizations** with toggle between views
-- **Color coding options** by category, subcategory, or tags
-- **Point inspection** - click points to view full document content
-- **Sidebar layout** with controls on left, large visualization area on right
-- **Real-time visualization** optimized for small to medium datasets
+**Important:** Document and prompt embeddings must have the same number of dimensions to be visualized together.
 
 ## Installation & Usage
 
@@ -68,7 +78,10 @@ uv run python app.py
 
 3. **Open your browser** to http://127.0.0.1:8050
 
-4. **Test with sample data** by dragging and dropping the included `sample_data.ndjson` file
+4. **Test with sample data**:
+   - Upload `sample_data.ndjson` (documents)
+   - Upload `sample_prompts.ndjson` (prompts) to see dual visualization
+   - Use the "Show prompts" toggle to compare how prompts relate to documents
 
 ## Tech Stack
 
diff --git a/embedding-buddy-screenshot.png b/embedding-buddy-screenshot.png
index 89185b7..8ecee66 100644
Binary files a/embedding-buddy-screenshot.png and b/embedding-buddy-screenshot.png differ