Initial commit with docker image ready for Hugging Face Spaces deployment

.gitignore

@@ -0,0 +1,3 @@
+.streamlit/secrets.toml
+*.env
+.env

.streamlit/config.toml

@@ -0,0 +1,4 @@
+[server]
+headless = true
+enableCORS = false
+port = 7860

Dockerfile

@@ -0,0 +1,25 @@
+# Base image with Python
+FROM python:3.10-slim
+
+# Set environment variable to prevent Python from writing .pyc files
+ENV PYTHONDONTWRITEBYTECODE=1
+
+# Set working directory inside the container
+WORKDIR /project
+
+# Copy all project files into the container
+COPY . .
+
+# Install system dependencies (you may need more later, like git or libgl)
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose the port that the app will run on
+EXPOSE 7860
+
+# Hugging Face expects this command to run the app
+CMD ["streamlit", "run", "app/main.py", "--server.port", "7860", "--server.address", "0.0.0.0"]

README.md

@@ -0,0 +1,3 @@
+# PromptAId Operations
+
+This is a simple chat prototype that will become an agentic RAG system for emergency operations. It currently uses a basic LLM pipeline. Future versions will support RAG, tool use, and knowledge graph querying.

app/main.py

@@ -0,0 +1,26 @@
+# app/main.py
+import streamlit as st  # For building the UI
+from langchain_openai import OpenAI  # OpenAI wrapper for LLM
+from langchain.chains import ConversationChain  # Simple chat pipeline
+import os
+openai_key = os.getenv("OPENAI_API_KEY")
+
+# Set Streamlit page config
+st.set_page_config(page_title="PromptAId Chat", page_icon="⚠️")
+
+# Display title
+st.title("🧠 PromptAId Operations — Chat Demo")
+
+# Input field for user prompt
+user_input = st.text_input("Ask something about risk or emergencies...")
+
+# Load your LLM (replace with your OpenAI API key in .streamlit/secrets.toml or env)
+llm = OpenAI(temperature=0.4)
+
+# Initialize a simple conversation chain (no memory yet)
+conversation = ConversationChain(llm=llm)
+
+# If there's input, run the LLM and display the result
+if user_input:
+    response = conversation.run(user_input)
+    st.markdown(f"**Model Response:** {response}")

requirements.txt

@@ -0,0 +1,5 @@
+streamlit
+openai
+langchain
+langchain-community
+langchain-openai