Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,178 @@
-import altair as alt
-import numpy as np
+import os
+import uuid
+import logging
 import pandas as pd
+import numpy as np
 import streamlit as st
+from huggingface_hub import InferenceClient
 
-"""
-# Welcome to Streamlit!
+# ======================
+# LOGGING
+# ======================
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger("AI-Agent")
+
+# ======================
+# CONFIG
+# ======================
+HF_MODEL = "mistralai/Mistral-7B-Instruct-v0.2"
+
+# ======================
+# LOAD DATA (500+ rows)
+# ======================
+@st.cache_data
+def load_data():
+    np.random.seed(42)
+    rows = 600
+    df = pd.DataFrame({
+        "order_id": range(1, rows + 1),
+        "supplier": np.random.choice(
+            ["Supplier A", "Supplier B", "Supplier C", "Supplier D"], rows
+        ),
+        "product": np.random.choice(
+            ["Tomato", "Cheese", "Flour", "Oil", "Meat"], rows
+        ),
+        "quantity": np.random.randint(1, 100, rows),
+        "price": np.random.uniform(5, 50, rows).round(2),
+        "delivery_days": np.random.randint(1, 14, rows),
+        "status": np.random.choice(
+            ["Delivered", "Pending", "Delayed"], rows
+        ),
+    })
+    logger.info("Loaded dataset with %d rows", len(df))
+    return df
+
+df = load_data()
+
+# ======================
+# TOOLS (FUNCTIONS)
+# ======================
+def query_database(product: str):
+    logger.info("Function call: query_database(%s)", product)
+    return df[df["product"].str.lower() == product.lower()].head(5)
+
+def get_aggregates():
+    logger.info("Function call: get_aggregates()")
+    return {
+        "rows": len(df),
+        "avg_price": round(df["price"].mean(), 2),
+        "avg_delivery_days": round(df["delivery_days"].mean(), 2),
+    }
+
+def create_support_ticket(text: str):
+    ticket_id = str(uuid.uuid4())[:8]
+    logger.info("Support ticket created: %s", ticket_id)
+    return {
+        "ticket_id": ticket_id,
+        "system": "GitHub Issues (mock)",
+        "status": "Created",
+        "description": text,
+    }
+
+# ======================
+# SAFETY CHECK
+# ======================
+def safety_guard(user_input: str):
+    blocked = ["delete", "drop", "truncate", "remove table"]
+    return any(b in user_input.lower() for b in blocked)
+
+# ======================
+# HF AGENT LOGIC
+# ======================
+def agent_response(user_input: str, client: InferenceClient):
+    if safety_guard(user_input):
+        logger.warning("Blocked dangerous request")
+        return "❌ This operation is not allowed."
+
+    # TOOL ROUTING (simple intent routing)
+    if "stats" in user_input or "summary" in user_input:
+        stats = get_aggregates()
+        return (
+            f"📊 Database Summary\n"
+            f"- Rows: {stats['rows']}\n"
+            f"- Avg Price: ${stats['avg_price']}\n"
+            f"- Avg Delivery Days: {stats['avg_delivery_days']}"
+        )
+
+    if user_input.lower().startswith("show"):
+        product = user_input.replace("show", "").strip()
+        result = query_database(product)
+        if result.empty:
+            return "No data found for that product."
+        return result
 
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+    if "support" in user_input or "ticket" in user_input:
+        ticket = create_support_ticket(user_input)
+        return (
+            f"🎫 Support Ticket Created\n"
+            f"- ID: {ticket['ticket_id']}\n"
+            f"- System: {ticket['system']}"
+        )
 
-In the meantime, below is an example of what you can do with just a few lines of code:
+    # HF LLM (TEXT ONLY, NO DB)
+    prompt = f"""
+You are a procurement assistant.
+Answer the user without assuming direct database access.
+
+User: {user_input}
+Assistant:
 """
+    logger.info("Calling HF model")
+
+    response = client.text_generation(
+        prompt,
+        max_new_tokens=200,
+        temperature=0.3,
+    )
+
+    return response
+
+# ======================
+# STREAMLIT UI
+# ======================
+st.set_page_config(page_title="AI Procurement Agent (HF)", layout="wide")
+st.title("🤖 AI Procurement Agent (Hugging Face Demo)")
+
+hf_token = st.sidebar.text_input(
+    "🔑 Hugging Face API Key",
+    type="password"
+)
+
+# Sidebar – business info
+st.sidebar.header("📊 Business Overview")
+stats = get_aggregates()
+st.sidebar.metric("Rows in DB", stats["rows"])
+st.sidebar.metric("Avg Price", f"${stats['avg_price']}")
+st.sidebar.metric("Avg Delivery Days", stats["avg_delivery_days"])
+
+st.sidebar.markdown("### Sample Queries")
+st.sidebar.code("""
+show tomato
+database stats
+create support ticket
+""")
+
+if not hf_token:
+    st.warning("Please enter Hugging Face API key to continue.")
+    st.stop()
+
+client = InferenceClient(
+    model=HF_MODEL,
+    token=hf_token
+)
+
+if "chat" not in st.session_state:
+    st.session_state.chat = []
+
+user_input = st.chat_input("Ask the agent...")
+
+if user_input:
+    st.session_state.chat.append(("user", user_input))
+    reply = agent_response(user_input, client)
+    st.session_state.chat.append(("assistant", reply))
 
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+for role, msg in st.session_state.chat:
+    if role == "user":
+        st.chat_message("user").write(msg)
+    else:
+        st.chat_message("assistant").write(msg)