Update configuration and documentation for improved clarity and functionality

- Added `api_key` field to `config.json` for better configuration management.
- Updated `default_model` in `config.json` to ensure consistency.
- Enhanced `README.md` with notes on model usage and API key management.
- Revised `api-reference.md` to clarify endpoint statuses and planned features.
- Updated `deployment.md` to reflect changes in application startup commands.
- Improved `faq.md` and `features.md` for better user guidance and feature clarity.

Author: TMHSDigital

README.md

@@ -119,6 +119,11 @@ Then open [http://localhost:8000](http://localhost:8000) in your browser.
 - Error banner for API issues
 - **Tool configuration in settings panel with API key management**
 
+Note on models:
+- Default documented model: `deepseek-chat`
+- You can change the runtime model via the REST API: `POST /api/set-model` (✓)
+- The UI Model tab is informational for now (X) and won’t switch the backend model yet
+
 For a comprehensive guide to using the web interface, see the [Web UI Guide](docs/web-ui-guide.md).
 
 ## Installation

config.json

@@ -1,6 +1,8 @@
 {
+  "api_key": null,
   "base_url": "https://api.deepseek.com/v1",
+  "default_model": "deepseek-chat",
   "timeout": 30.0,
   "max_retries": 3,
-  "default_model": "deepseek-chat"
+  "extract_answer_only": false
 }

docs/api-reference.md

@@ -361,213 +361,59 @@ except DeepSeekAuthError as e:
 
 ## REST API Endpoints
 
-The DeepSeek Wrapper provides a FastAPI-based REST API with the following endpoints:
+This section documents the actual FastAPI endpoints exposed by the app and marks planned endpoints for future releases.
+
+### Status legend
+- ✓ Implemented
+- X Planned (not implemented yet)
+
+### Endpoint summary
+
+| Status | Method | Path |
+|--------|--------|------|
+| ✓ | GET | `/` |
+| ✓ | POST | `/chat` |
+| ✓ | POST | `/completions` |
+| ✓ | POST | `/reset` |
+| ✓ | POST | `/upload` |
+| ✓ | GET | `/chat/stream` |
+| ✓ | POST | `/save_api_keys` |
+| ✓ | POST | `/api/save-api-keys` |
+| ✓ | GET | `/api/key-status` |
+| ✓ | GET | `/api/tool-status` |
+| ✓ | POST | `/api/clear-caches` |
+| ✓ | GET | `/api/test-weather` |
+| ✓ | GET | `/test-weather` |
+| ✓ | GET | `/api/models` |
+| ✓ | POST | `/api/set-model` |
+| ✓ | GET | `/api/config` |
+| ✓ | POST | `/api/set-config` |
+| X | POST | `/api/chat` |
+| X | POST | `/api/generate` |
+| X | GET | `/api/generate/stream` |
+| X | POST | `/api/embeddings` |
+| X | GET/POST/DELETE | `/api/conversations*` |
+| X | GET/POST/DELETE | `/api/webhooks*` |
+
+### Streaming chat (✓ GET `/chat/stream`)
+
+Server-Sent Events stream of assistant output. Message types:
 
-### Chat Endpoints
-
-#### `POST /api/chat`
-
-Create a new chat message and get a response from the DeepSeek AI.
-
-**Request Body:**
-```json
-{
-  "message": "Your message here",
-  "conversation_id": "optional-conversation-id",
-  "system_prompt": "Optional system prompt to guide the AI's behavior"
-}
-```
-
-**Response:**
-```json
-{
-  "id": "response-id",
-  "content": "AI response content",
-  "conversation_id": "conversation-id"
-}
-```
-
-#### `GET /api/chat/stream`
-
-Stream a chat response using Server-Sent Events (SSE).
-
-**Query Parameters:**
-- `message` (string): The user's message
-- `conversation_id` (string, optional): Conversation identifier
-- `system_prompt` (string, optional): System prompt
-
-**Response:**
-Server-sent events with the following data format:
 ```json
-{
-  "id": "chunk-id",
-  "content": "Partial content chunk",
-  "is_complete": false
-}
+{ "type": "user_msg_received", "message": {"role":"user","content":"..."} }
+{ "type": "assistant_msg_start", "message": {"role":"assistant","content":""} }
+{ "type": "content_chunk", "chunk": "partial text" }
+{ "type": "replace_content", "content": "final processed text" }
+{ "type": "complete", "message": {"role":"assistant","content":"..."} }
+{ "type": "error", "error": "message" }
 ```
 
-Final chunk will have `is_complete` set to `true`.
-
-### Text Generation Endpoints
-
-#### `POST /api/generate`
-
-Generate text based on a prompt without conversation context.
-
-**Request Body:**
-```json
-{
-  "prompt": "Your prompt here",
-  "max_tokens": 100,
-  "temperature": 0.7
-}
-```
-
-**Response:**
-```json
-{
-  "id": "generation-id",
-  "text": "Generated text content"
-}
-```
-
-#### `GET /api/generate/stream`
-
-Stream a text generation response using Server-Sent Events (SSE).
-
-**Query Parameters:**
-- `prompt` (string): The generation prompt
-- `max_tokens` (integer, optional): Maximum tokens to generate
-- `temperature` (float, optional): Temperature for generation
-
-**Response:**
-Server-sent events with the same format as the chat stream endpoint.
-
-### Embedding Endpoints
-
-#### `POST /api/embeddings`
-
-Generate embeddings for input text.
-
-**Request Body:**
-```json
-{
-  "input": "Text to generate embeddings for"
-}
-```
-
-**Response:**
-```json
-{
-  "embedding": [0.123, 0.456, ...], // Vector of floating-point values
-  "dimensions": 1536 // Number of dimensions in the embedding vector
-}
-```
-
-### Conversation Endpoints
-
-#### `GET /api/conversations`
-
-Get a list of all conversations.
-
-**Response:**
-```json
-[
-  {
-    "id": "conversation-id",
-    "title": "Conversation title",
-    "created_at": "2023-06-15T14:30:00Z",
-    "updated_at": "2023-06-15T14:35:00Z"
-  }
-]
-```
-
-#### `GET /api/conversations/{conversation_id}`
-
-Get details and messages for a specific conversation.
-
-**Response:**
-```json
-{
-  "id": "conversation-id",
-  "title": "Conversation title",
-  "created_at": "2023-06-15T14:30:00Z",
-  "updated_at": "2023-06-15T14:35:00Z",
-  "messages": [
-    {
-      "id": "message-id",
-      "role": "user",
-      "content": "User message",
-      "created_at": "2023-06-15T14:30:00Z"
-    },
-    {
-      "id": "response-id",
-      "role": "assistant",
-      "content": "AI response",
-      "created_at": "2023-06-15T14:31:00Z"
-    }
-  ]
-}
-```
-
-#### `DELETE /api/conversations/{conversation_id}`
-
-Delete a specific conversation.
-
-**Response:** Status 204 No Content
-
-### Document Endpoints
-
-#### `POST /api/documents/upload`
-
-Upload a document for processing.
-
-**Request:** Multipart form data with a `file` field containing the document.
-
-**Response:**
-```json
-{
-  "id": "document-id",
-  "filename": "document.pdf",
-  "content": "Extracted text from the document",
-  "content_type": "application/pdf",
-  "size": 1024
-}
-```
-
-### Webhook Endpoints
-
-#### `POST /api/webhooks/register`
-
-Register a webhook to receive notifications.
-
-**Request Body:**
-```json
-{
-  "url": "https://your-webhook-endpoint.com",
-  "events": ["completion.finished", "error"],
-  "description": "Optional description of this webhook"
-}
-```
-
-**Response:**
-```json
-{
-  "id": "webhook-id",
-  "url": "https://your-webhook-endpoint.com",
-  "events": ["completion.finished", "error"],
-  "created_at": "2023-06-15T14:30:00Z"
-}
-```
-
-#### `GET /api/webhooks`
-
-List all registered webhooks.
-
-**Response:** Array of webhook objects.
-
-#### `DELETE /api/webhooks/{webhook_id}`
+Query params:
+- `user_message`: required
+- `system_prompt`: optional
 
-Delete a registered webhook.
+### Chat (✓ POST `/chat`)
+Form field `user_message` with current session history; returns the rendered HTML page.
 
-**Response:** Status 204 No Content 
+### Completions (✓ POST `/completions`)
+Form field `prompt`; returns the rendered HTML page with result.

docs/deployment.md

@@ -9,7 +9,7 @@ This guide covers different options for deploying the DeepSeek Wrapper applicati
 For local development, you can run the application with:
 
 ```bash
-python -m src.deepseek_wrapper.main
+uvicorn src.deepseek_wrapper.web:app --reload
 ```
 
 This will start the application with auto-reload enabled for development purposes.
@@ -19,7 +19,7 @@ This will start the application with auto-reload enabled for development purpose
 For a more production-ready local deployment:
 
 ```bash
-uvicorn src.deepseek_wrapper.main:app --host 0.0.0.0 --port 8000 --workers 4
+uvicorn src.deepseek_wrapper.web:app --host 0.0.0.0 --port 8000 --workers 4
 ```
 
 This uses the Uvicorn ASGI server with multiple worker processes for better performance.
@@ -126,11 +126,17 @@ The following environment variables can be configured for deployment:
 | Variable | Description | Default |
 |----------|-------------|---------|
 | `DEEPSEEK_API_KEY` | Your DeepSeek API key (required) | None |
-| `HOST` | Host to bind the server to | 0.0.0.0 |
-| `PORT` | Port to run the server on | 8000 |
-| `LOG_LEVEL` | Logging level (DEBUG, INFO, WARNING, ERROR) | INFO |
-| `UPLOAD_DIR` | Directory for file uploads | ./uploads |
-| `MAX_UPLOAD_SIZE` | Maximum file upload size in MB | 10 |
+| `DEEPSEEK_BASE_URL` | API base URL | https://api.deepseek.com/v1 |
+| `DEEPSEEK_DEFAULT_MODEL` | Default model name | deepseek-chat |
+| `SESSION_SECRET_KEY` | Secret for session middleware | supersecret |
+| `SEARCH_API_KEY` | API key for Web Search tool |  |
+| `OPENWEATHERMAP_API_KEY` | API key for Weather tool |  |
+| `EMAIL_SMTP_SERVER` | SMTP server for Email tool | smtp.gmail.com |
+| `EMAIL_SMTP_PORT` | SMTP port | 587 |
+| `EMAIL_USERNAME` | SMTP username |  |
+| `EMAIL_PASSWORD` | SMTP password |  |
+| `EMAIL_TEMPLATE_DIR` | Email templates directory | ./email_templates |
+| `WOLFRAM_ALPHA_APP_ID` | App ID for Wolfram Alpha tool |  |
 
 ## Scaling Considerations
 

docs/faq.md

@@ -34,7 +34,7 @@ No, the wrapper needs to connect to the DeepSeek API servers, which requires an
 ## Usage
 
 ### How do I start a new conversation?
-In the web UI, click the "New Chat" button in the sidebar. In the API, make a request without specifying a conversation ID to start a new one.
+In the web UI, click the "New Chat" button in the sidebar. Programmatically, post to `/chat` with a `user_message` form field to begin a session-backed exchange.
 
 ### Can I upload files to provide context?
 Yes, you can upload PDF, DOCX, and TXT files using the file upload button next to the input field. The content will be processed and included in the context for your next message.
@@ -43,7 +43,7 @@ Yes, you can upload PDF, DOCX, and TXT files using the file upload button next t
 In the web UI, click the settings icon and enter your system prompt in the appropriate field. In the API, include the `system_prompt` parameter in your request.
 
 ### Can I export my conversations?
-Currently, there's no built-in export feature, but this is planned for a future release.
+Currently, there's no built-in export feature. X Planned for a future release.
 
 ## Troubleshooting
 
@@ -76,10 +76,10 @@ See the [CONTRIBUTING.md](CONTRIBUTING.md) file for guidelines on how to contrib
 Yes, the frontend is built with HTML, CSS, and JavaScript. You can modify or extend it by editing the templates and static files.
 
 ### How do I add support for a new file format?
-To add support for a new file format, you'll need to implement a parser in the `document_processor.py` file and register it in the file type handlers.
+Add support in `src/deepseek_wrapper/web.py` within `extract_text_from_file` to parse the new type, and extend upload handling if needed.
 
 ### Is there an API for programmatic access?
-Yes, you can use the Python client API or make HTTP requests directly to the REST API endpoints. See the [API Reference](api-reference.md) for details.
+Yes. Use the Python client API, or the implemented FastAPI endpoints documented in the [API Reference](api-reference.md). Some broader REST endpoints are X planned.
 
 ## Pricing & Limits
 

docs/features.md

@@ -30,9 +30,9 @@
   <div style="flex: 1; min-width: 300px; border-left: 4px solid #FF6B6B; padding: 10px; background: #FFF5F5; margin: 5px 0;">
     <h3 style="color: #FF6B6B; margin-top: 0;">Production Ready</h3>
     <ul>
-      <li><b>Comprehensive logging</b></li>
-      <li><b>Request retry logic</b></li>
-      <li><b>Auto-scaling handling</b></li>
+      <li><b>Comprehensive logging</b> (✓)</li>
+      <li><b>Request retry logic</b> (✓)</li>
+      <li><b>Auto-scaling handling</b> (X via infrastructure)</li>
     </ul>
   </div>
 </div>
@@ -137,13 +137,13 @@
   <div style="flex: 1; min-width: 300px;">
     <h3 style="color: #FF6B6B; border-bottom: 2px solid #FF6B6B; padding-bottom: 5px;">Security</h3>
     <ul>
-      <li>Environment-based configuration for secure API key storage</li>
-      <li>Input validation and sanitization to prevent injection attacks</li>
-      <li>Secure file handling for uploads with size and type validation</li>
-      <li>HTTPS-only communications with the DeepSeek API</li>
-      <li>Rate limiting to prevent abuse</li>
-      <li>Content validation to filter potentially harmful outputs</li>
-      <li>Automatic session expiration for inactive users</li>
+      <li>Environment-based configuration for secure API key storage (✓)</li>
+      <li>Input validation and sanitization to prevent injection attacks (partial ✓)</li>
+      <li>Secure file handling for uploads with size and type validation (partial ✓)</li>
+      <li>HTTPS-only communications with the DeepSeek API (X — configure at deployment/proxy)</li>
+      <li>Rate limiting to prevent abuse (X overall; Weather tool has internal limiter ✓)</li>
+      <li>Content validation to filter potentially harmful outputs (X)</li>
+      <li>Automatic session expiration for inactive users (X)</li>
     </ul>
   </div>
 </div>