ai-support-intelligence-hub/app_pro_final.py at main · joicyroslin-svg/ai-support-intelligence-hub · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
from __future__ import annotations

import json
import re
import time
from pathlib import Path
from typing import Any, Dict, List

import streamlit as st
import streamlit_confetti as st_confetti
import plotly.graph_objects as go
# import pandas as pd  # removed for policy

from assistant.config import DEFAULT_MODEL, get_available_models, validate_keys
from assistant.llm import LiteLLMClient
from assistant.logging_utils import LOG_FILE
from assistant.rag import auto_seed_rag, rag, load_csv_to_rag, get_all_tickets_df
from assistant.service import analyze_ticket, build_agent_service, bulk_analyze_tickets
from assistant.agent import run_agent
from assistant.tools import TOOLS
from assistant.ui_theme import apply_figma_theme

st.set_page_config(page_title="AI Support Pro", layout="wide")

# Fallbacks for optional

PLOTLY_AVAILABLE = True


def safe_render_table(data: Any, max_rows: int = 100) -> None:
    """Render tabular data without requiring pyarrow."""
    if data is None:
        st.info("No data available.")
        return

    try:
        if hasattr(data, "empty") and getattr(data, "empty"):
            st.info("No data available.")
            return
    except Exception:
        pass

    rows: list[Any] = []

    try:
        if hasattr(data, "to_dict"):
            maybe_rows = data.to_dict(orient="records")
            rows = maybe_rows if isinstance(maybe_rows, list) else [maybe_rows]
        elif isinstance(data, list):
            rows = data
        else:
            rows = [data]
    except Exception:
        rows = [data]

    if not rows:
        st.info("No data available.")
        return

    for idx, row in enumerate(rows[:max_rows], start=1):
        if isinstance(row, dict):
            with st.expander(f"Row {idx}", expanded=idx <= 5):
                for key, value in row.items():
                    st.write(f"**{key}:** {value}")
        else:
            st.write(row)

    if len(rows) > max_rows:
        st.caption(f"Showing first {max_rows} rows of {len(rows)} total.")


def inject_styles():
    st.markdown('''
<style>
@import url('https://fonts.googleapis.com/css2?family=Manrope:wght@400;600;700;800&family=Space+Grotesk:wght@600;700&display=swap');
:root {
  --bg: #f8fbff;
  --panel: #ffffff;
  --text: #102540;
  --border: #d5e4f7;
}
[data-theme="dark"] {
  --bg: #0f172a;
  --panel: #1e293b;
  --text: #f1f5f9;
  --border: #334155;
}
body { font-family: 'Manrope', sans-serif; }
.hero { background: linear-gradient(135deg, var(--panel), #eef6ff); border-radius: 20px; padding: 2rem; box-shadow: 0 10px 30px rgba(0,0,0,0.1); border: 1px solid var(--border); }
.kpi { background: var(--panel); border-radius: 12px; padding: 1rem; border: 1px solid var(--border); box-shadow: 0 4px 10px rgba(0,0,0,0.05); }
.stButton > button { background: linear-gradient(90deg, #1266eb, #16aa82) !important; color: white !important; border-radius: 12px !important; }
.panel { background: var(--panel); border-radius: 12px; padding: 1.5rem; border: 1px solid var(--border); margin: 1rem 0; }
.tag { background: #f0f8ff; color: #1e40af; padding: 0.25rem 0.5rem; border-radius: 20px; font-size: 0.8rem; font-weight: 600; border: 1px solid #bfdbfe; margin: 0.25rem; display: inline-block; }
</style>
    ''', unsafe_allow_html=True)

inject_styles()
apply_figma_theme(
    "🚀 AI Support Pro Dashboard",
    "Professional, error-free AI triage with a cleaner Figma-style presentation layer.",
    "Support workspace",
)

st.session_state.setdefault('history', [])
st.session_state.setdefault('ticket_text', '')
st.session_state.setdefault('reply_draft', '')
st.session_state.setdefault('last_result', {})

with st.sidebar:
    st.header("Controls")
    theme = st.selectbox("Theme", ["Light", "Dark"])
    st.markdown(f'<body data-theme="{theme.lower()}"></body>', unsafe_allow_html=True)

    if validate_keys().get("HF_TOKEN"):
        st.success("✅ HF Active")
    else:
        st.info("Free mode")

    model = st.selectbox("Model", get_available_models() or [DEFAULT_MODEL])
    depth = st.slider("Creativity", 0.0, 0.5, 0.2)

    col_a, col_b = st.columns(2)
    if col_a.button("Seed Demo Data", use_container_width=True):
        st.success(f"Seeded {auto_seed_rag(rag, 50)} tickets")

    csv_path = "data/sample_tickets.csv"
    if col_b.button("Load CSV", use_container_width=True):
        loaded = load_csv_to_rag(csv_path, rag)
        st.success(f"Loaded {loaded} tickets from CSV!")

    if st.button("Clear Session", use_container_width=True):
        st.session_state.history = []
        st.session_state.ticket_text = ""
        st.session_state.reply_draft = ""
        st.session_state.chat_history = []
        st.session_state.last_result = {}
        st.success("Session cleared.")

tab1, tab2, tab3, tab4 = st.tabs(["Tickets", "Build Agent", "Realtime Chat", "Bulk Analysis"])


with tab3:
    st.header("🤖 Realtime Chat Agent")
    st.caption("Use this tab for quick support-copilot answers based on the selected model.")
    st.session_state.setdefault("chat_history", [])
    st.session_state.setdefault("chat_input", "")

    for msg in st.session_state.chat_history:
        with st.chat_message(msg["role"]):
            st.write(msg["content"])

    if prompt := st.chat_input("Ask the support agent..."):
        st.session_state.chat_history.append({"role": "user", "content": prompt})
        with st.chat_message("user"):
            st.write(prompt)

        with st.chat_message("assistant"):
            with st.spinner("Thinking..."):
                response = run_agent(prompt, model)
            st.write(response["response"])
            if response.get("error"):
                st.caption("Fallback response shown because the model was unavailable.")

        st.session_state.chat_history.append({"role": "assistant", "content": response["response"]})


with tab4:
    st.markdown('<div class="glass-card">', unsafe_allow_html=True)
    st.header("📊 Bulk Analysis")
    tickets_df = get_all_tickets_df(rag)

    if tickets_df:
        st.caption(f"{len(tickets_df)} tickets available for bulk analysis.")
        safe_render_table(tickets_df)
    else:
        st.info("No tickets available yet. Use Seed Demo Data or Load CSV first.")

    if st.button("Analyze All Tickets", use_container_width=True):
        if tickets_df:
            with st.spinner("Bulk analyzing..."):
                analyses = bulk_analyze_tickets(tickets_df)
            safe_render_table(analyses)

            pri_count = {}
            cat_count = {}
            for analysis in analyses:
                pri = analysis.get("priority", "Unknown")
                cat = analysis.get("category", "Unknown")
                pri_count[pri] = pri_count.get(pri, 0) + 1
                cat_count[cat] = cat_count.get(cat, 0) + 1

            fig_pri = go.Figure(data=[go.Pie(labels=list(pri_count.keys()), values=list(pri_count.values()))])
            fig_pri.update_layout(title="Priority Distribution")
            st.plotly_chart(fig_pri, use_container_width=True)

            fig_cat = go.Figure(data=[go.Bar(x=list(cat_count.keys()), y=list(cat_count.values()))])
            fig_cat.update_layout(title="Categories")
            st.plotly_chart(fig_cat, use_container_width=True)
        else:
            st.warning("Add ticket data before running bulk analysis.")
    st.markdown('</div>', unsafe_allow_html=True)

with tab1:
    st.markdown('<div class="glass-card">', unsafe_allow_html=True)
    if st.button("Payment Issue", use_container_width=True):
        st.session_state.ticket_text = "Payment failed twice, worried about charge."

    ticket = st.text_area(
        "Support Ticket",
        value=st.session_state.ticket_text,
        height=200,
        placeholder="Paste the customer message here for analysis...",
    )

    col1, col2 = st.columns(2)
    tone = col1.selectbox("Tone", ["professional", "empathetic"])
    language = col2.selectbox("Language", ["English"])

    if st.button("Analyze Ticket", use_container_width=True):
        if ticket.strip():
            try:
                with st.spinner("AI Processing..."):
                    client = LiteLLMClient(model)
                    result, _, _ = analyze_ticket(
                        ticket,
                        client,
                        tone,
                        language,
                        include_followups=True,
                        include_internal_notes=True,
                        include_tags=True,
                        redact=True,
                        temperature=depth,
                        max_output_tokens=1000,
                    )
                st.session_state.last_result = result
                st.session_state.ticket_text = ticket
                st.session_state.reply_draft = result.get("reply", "")
                st.session_state.history.append(result)
                st.success("✅ Analysis Complete")
                st.caption("Priority, category, sentiment, reply draft, and history charts updated below.")
            except Exception as e:
                st.error(f"Failed: {e}")
        else:
            st.warning("Enter ticket to analyze")

    last_result = st.session_state.get('last_result')
    if last_result is not None and isinstance(last_result, dict) and last_result:
        col1, col2, col3, col4 = st.columns(4)
        col1.metric("Priority", last_result.get('priority', '?'))
        col2.metric("Category", last_result.get('category', '?'))
        col3.metric("Sentiment", last_result.get('sentiment', '?'))
        col4.metric("Reply Ready", "Yes")

        action_col1, action_col2 = st.columns(2)
        with action_col1:
            st.download_button(
                "Download Analysis JSON",
                data=json.dumps(last_result, indent=2),
                file_name="ticket_analysis.json",
                mime="application/json",
                use_container_width=True,
            )
        with action_col2:
            st.download_button(
                "Download Reply TXT",
                data=st.session_state.reply_draft,
                file_name="reply_draft.txt",
                mime="text/plain",
                use_container_width=True,
            )

        c1, c2 = st.columns(2)
        with c1:
            st.markdown('<div class="panel">', unsafe_allow_html=True)
            st.markdown("**Reply:**")
            st.text_area("Reply Draft", value=st.session_state.reply_draft, height=200)
            st.markdown('</div>', unsafe_allow_html=True)
        with c2:
            st.markdown('<div class="panel">', unsafe_allow_html=True)
            st.markdown("**Insights:**")
            for k, v in last_result.items():  # type: ignore[union-attr]
                st.markdown(f"**{k}:** {v}")
            st.markdown('</div>', unsafe_allow_html=True)

        if st.session_state.history:
            st.subheader("History Summary")
            pri_count = {}
            cat_count = {}
            for item in st.session_state.history:
                pri = item.get('priority', 'Unknown')
                cat = item.get('category', 'Unknown')
                pri_count[pri] = pri_count.get(pri, 0) + 1
                cat_count[cat] = cat_count.get(cat, 0) + 1

            fig_pri = go.Figure(data=[go.Pie(labels=list(pri_count.keys()), values=list(pri_count.values()))])
            fig_pri.update_layout(title="History Priority")
            st.plotly_chart(fig_pri, use_container_width=True)

            fig_cat = go.Figure(data=[go.Bar(x=list(cat_count.keys()), y=list(cat_count.values()))])
            fig_cat.update_layout(title="History Categories")
            st.plotly_chart(fig_cat, use_container_width=True)

            st.download_button(
                "Download History JSON",
                data=json.dumps(st.session_state.history, indent=2),
                file_name="analysis_history.json",
                mime="application/json",
                use_container_width=True,
            )
    st.markdown('</div>', unsafe_allow_html=True)

with tab2:
    st.markdown('<div class="glass-card">', unsafe_allow_html=True)
    st.header("Agent Builder")
    st.caption("Generate a starter support-agent config from the tools and goal you provide.")
    tools_input = st.text_input("Tools (comma separated)", "chatgpt, tools")
    goal = st.text_area("Goal", "Support copilot")
    if st.button("Generate Agent", use_container_width=True):
        if goal.strip():
            try:
                tools_list = [t.strip() for t in tools_input.split(",") if t.strip()]
                config = build_agent_service(tools_list, goal, model)
                st.success("Agent configuration generated.")
                st.json(config)
            except Exception as e:
                st.error(str(e))
        else:
            st.warning("Enter an agent goal before generating.")
    st.markdown('</div>', unsafe_allow_html=True)

st.markdown('<p class="footer-note">✅ Professional, error-free AI dashboard with safe Figma-style polish applied.</p>', unsafe_allow_html=True)