created a test script to verify the MCP endpoint

include MCP-specific configurations

interactive questioning enhancement

getting local to work to prevent burining up my credits

neaten

hf embeddings for llama index had to be installed: https://docs.llamaindex.ai/en/stable/module_guides/models/embeddings/

If you’re using Gradio’s built-in browser mic input (e.g. gr.Audio(source="microphone") for real-time ASR), you can drop both sounddevice and pyaudio. You’ll still need ffmpeg on the server (or your dev box) so Gradio can transcode incoming audio blobs

removed unused imports

fix: ImportError: cannot import name 'SimpleDirectoryReader' from 'llama_index' (unknown location)

switching from openaid whisper to gradio asr: https://www.gradio.app/guides/real-time-speech-recognition

readme is good now

Merge remote-tracking branch 'competition/main'

Merge remote-tracking branch 'competition/main'

initial commit

Duplicate from gradio-templates/chatbot

added a streaming Audio input component per the Gradio guide—using Whisper for transcription in real time and piping the resulting text into the chatflow

In Gradio Spaces (and all HF SDKs), any Debian-level dependencies go in a file called packages.txt at the root of your repo

Add apt.txt for system dependencies (portaudio, ffmpeg, etc.) and remove Dockerfile

chore: add Dockerfile to install PortAudio, FFmpeg, and image libs for Space builds

Add optional audio dependencies and FFmpeg note to requirements.txt

description was too long

invalid tag

Update README to reflect audio-driven pipeline and multi-backend LLM support

Add langchain and langchain-community to requirements

Add multi‐backend LLM support and audio‐driven medical agent pipeline

Use `demo` as Gradio entrypoint identifier

Use `demo` as Gradio entrypoint identifier

mcp_server=true came from chatgpt, isn't valid

using the wrapper is way better because it sets the pwd to the root of the repo

Add root-level app.py wrapper; preserve src/ layout

using gradio 5.33.0

added ability to develop on my local rtx2060

utilizing llama index can net an extra $1000 in winnings

interactive questioning is key and demos the unique capabilities of ai

my current goal is to get a simple version of the Symptom-to-ICD MCP server running

mvp checklist

tag is required for competition submission

badly needed link added

icd system isn't actually that useful

finished reading Strategy to Win the MCP Tool/Server Hackathon Track paper

MedCodeMCP is clearer. Capitalizing the “C” separates “Med” and “Code” and avoids the run-together feel of MedcodeMCP

readme file from medicode repo is the only valuable file there, will delete repo

repo structure cleanup (https://chatgpt.com/share/6841a61b-3dfc-8004-aec7-9c48fba857fd)

3 relevant papers so far

included name of papers I have already

chatgpt research papers will be essential

strategic research papers

ICD-10 coding and confidence scoring

successfully scrapped

finalized my winning idea for this hackathon, starting to build my knowledge base file based on ICD10