have to turn the mcp server on for it to work

README.md

@@ -14,13 +14,14 @@ tags:
   - agent-demo-track
 ---
 
-A voice-enabled medical assistant that takes patient text or audio complaints, engages in follow-up questions, and returns structured ICD-10 diagnosis suggestions via an MCP endpoint.
+MedCodeMCP is a voice-enabled medical diagnosis assistant that maps patient symptoms to probable ICD-10 codes and explanations. It leverages automatic speech recognition (ASR) (via Whisper) to transcribe spoken complaints and then engages an interactive Q&A session with the user to clarify symptoms.
+
+Under the hood, it uses a Large Language Model (LLM) to either ask focused follow-up questions or output a final diagnosis suggestion in JSON format (with fields for diagnoses and confidence scores). The application integrates a vector database of ICD-10 descriptions using LlamaIndex for retrieval: given symptoms, it finds relevant ICD-10 code information to inform the LLM’s reasoning.
 
 # Features
 
 - **Automatic speech recognition (ASR)**: Transcribe real-time patient audio using [Gradio](https://www.gradio.app/guides/real-time-speech-recognition).
 - **Interactive Q&A Agent**: The LLM dynamically asks clarifying questions base on ICD codes until it can diagnose with high confidence.
-- **Multi-backend LLM**: Switch between OpenAI GPT, Mistral (HF), or a local transformers model via environment flags.
 - **ICD-10 Mapping**: Use LlamaIndex for vector retrieval of probable ICD-10 codes with confidence scores.
 - **MCP-Server Ready**: Exposes a `/mcp` REST endpoint for seamless agent integration.
 
@@ -53,39 +54,10 @@ git clone https://huggingface.co/spaces/Agents-MCP-Hackathon/MedCodeMCP
 cd MedCodeMCP
 python3 -m venv .venv && source .venv/bin/activate  
 pip install -r requirements.txt
-````
-
-## Environment Variables
-
-| Name                       | Description                                               | Default                |
-| -------------------------- | --------------------------------------------------------- | ---------------------- |
-| `OPENAI_API_KEY`           | OpenAI API key for GPT calls                              | *required*             |
-| `HUGGINGFACEHUB_API_TOKEN` | HF token for Mistral/inference models                     | *required for Mistral* |
-| `USE_LOCAL_GPU`            | Set to `1` to use a local transformers model (no credits) | `0`                    |
-| `LOCAL_MODEL`              | Path or HF ID of local model (e.g. `distilgpt2`)          | `gpt2`                 |
-| `USE_MISTRAL`              | Set to `1` to use Mistral via HF instead of OpenAI        | `0`                    |
-| `MISTRAL_MODEL`            | HF ID for Mistral model (`mistral-small/medium/large`)    | `mistral-large`        |
-| `MISTRAL_TEMPERATURE`      | Sampling temperature for Mistral                          | `0.7`                  |
-| `MISTRAL_MAX_INPUT`        | Max tokens for input prompt                               | `4096`                 |
-| `MISTRAL_NUM_OUTPUT`       | Max tokens to generate                                    | `512`                  |
+````     |
 
 ## Launch Locally
 
-```bash
-# Default (OpenAI)
-python app.py
-
-# Mistral backend
-export USE_MISTRAL=1
-export HUGGINGFACEHUB_API_TOKEN="hf_..."
-python app.py
-
-# Local GPU (no credits)
-export USE_LOCAL_GPU=1
-export LOCAL_MODEL="./models/distilgpt2"
-python app.py
-```
-
 Open [http://localhost:7860](http://localhost:7860) to interact with the app.
 
 ## MCP API Usage

app.py

@@ -190,7 +190,6 @@ with gr.Blocks(theme="default") as demo:
         fn=update_live_transcription,
         inputs=[microphone],
         outputs=[text_input],
-        show_progress=False,
         queue=True
     )
 
@@ -215,4 +214,4 @@ with gr.Blocks(theme="default") as demo:
     )
 
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860, share=True, show_api=True)
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=True, show_api=True, mcp_server=True)