Files
HOALedgerIQ_Website/agents/junior-ae/junior-ae-v3.py
olsch01 5319bcd30b feat: Add Chatwoot Agent Bot prototype and FAQ knowledge base
- Created chatwoot-agent-bot/ with Node.js webhook server
- Bot detects intent (greeting, billing, technical, features, account)
- Auto-responds from FAQ knowledge base or escalates to human
- FAQ-KB.md: Living knowledge base that grows with customer questions
- CHATWOOT-SETUP.md: Complete deployment and configuration guide
- Supports Telegram notifications on escalation
- Bot runs on port 3001, ready for Chatwoot webhook integration
2026-04-01 16:26:05 -04:00

195 lines
6.7 KiB
Python

#!/usr/bin/env python3
"""Junior AE v3 - Updates CRM Temp field directly"""
import json, re, time, urllib.request, urllib.error
from datetime import datetime, timedelta
from pathlib import Path
import ssl
SCRIPT_DIR = Path(__file__).parent
for d in [SCRIPT_DIR / "state", SCRIPT_DIR / "logs"]:
d.mkdir(parents=True, exist_ok=True)
STATE_FILE = SCRIPT_DIR / "state" / "jae-v3-state.json"
LOG_FILE = SCRIPT_DIR / "logs" / f"jae-v3-{datetime.now().strftime('%Y%m%d')}.log"
CRM_URL = "https://salesforce.hoaledgeriq.com/rest"
CRM_TOKEN = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiI5M2FmNGFmNS0zZWQ0LTQ1ZDMtOWE5Zi01MDMzZjc3YTY3MjMiLCJ0eXBlIjoiQVBJX0tFWSIsIndvcmtzcGFjZUlkIjoiOTNhZjRhZjUtM2VkNC00NWQzLTlhOWYtNTAzM2Y3N2E2NzIzIiwiaWF0IjoxNzczMzI4NDQzLCJleHAiOjE4MDQ3ODE2NDIsImp0aSI6IjIwZjEyYzkwLTRkMDctNGJmNi1iMzk3LTZjNmU3MzlmMThjOCJ9.zeM5NvwCSGEcz99m2LYtgb0sVD6WUXcCF7SwonFg930"
def log(msg):
ts = datetime.now().strftime('%H:%M:%S')
print(f"[{ts}] {msg}")
with open(LOG_FILE, 'a') as f:
f.write(f"[{ts}] {msg}\n")
def load_state():
if STATE_FILE.exists():
return json.loads(STATE_FILE.read_text())
return {"last_check": (datetime.now() - timedelta(days=7)).isoformat(), "processed": 0, "upgraded": 0}
def save_state(s):
STATE_FILE.write_text(json.dumps(s, indent=2))
def fetch_notes():
try:
req = urllib.request.Request(
f"{CRM_URL}/notes?limit=100&order[createdAt]=desc",
headers={"Authorization": f"Bearer {CRM_TOKEN}", "Accept": "application/json"}
)
with urllib.request.urlopen(req, timeout=15) as r:
return json.loads(r.read().decode()).get('data', {}).get('notes', [])
except Exception as e:
log(f"Fetch error: {e}")
return []
def get_temp(title):
t = title.upper()
if 'HOT' in t: return 'HOT'
if 'WARM' in t: return 'WARM'
if 'COLD' in t: return 'COLD'
return None
def extract_url(body):
if not body:
return None
m = re.search(r'Site:\s*(https?://[^\s\n<]+)', str(body))
if m:
return m.group(1).strip()
m = re.search(r'(https?://[^\s\n<"]+)', str(body))
return m.group(1) if m else None
def validate_website(url):
if not url:
return False, "no_url"
if not url.startswith('http'):
url = 'https://' + url
try:
ssl_context = ssl._create_unverified_context()
req = urllib.request.Request(
url,
headers={
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)",
"Accept": "text/html,*/*",
}
)
with urllib.request.urlopen(req, timeout=15, context=ssl_context) as r:
content = r.read()
code = r.getcode()
if code != 200:
return False, f"http_{code}"
if len(content) < 500:
return False, "too_small"
html = content.decode('utf-8', errors='ignore')[:3000].lower()
has_title = '<title>' in html
has_body = '<body' in html
text = re.sub(r'<[^>]+>', '', html)
has_content = len(text.strip()) > 100
if has_title and has_body and has_content:
return True, "real_website"
return False, f"missing_{'title' if not has_title else 'body' if not has_body else 'content'}"
except urllib.error.HTTPError as e:
if e.code in [301, 302]:
new_url = e.headers.get('Location', '')
if new_url and new_url != url:
return validate_website(new_url)
return False, f"http_{e.code}"
except Exception as e:
return False, str(e)[:30]
def upgrade(temp):
return {'COLD': 'WARM', 'WARM': 'HOT', 'HOT': 'HOT'}.get(temp, temp)
def update_note_full(note_id, body, title, new_temp, status):
"""Update CRM Temp field and title"""
try:
# Update title to reflect new temp
new_title = title
for old in ['COLD', 'WARM', 'HOT']:
if old in title:
new_title = title.replace(old, new_temp)
break
# Update body with validation info
new_body = body + f"\n\n**JAE v3:** {datetime.now().strftime('%Y-%m-%d %H:%M')}\n" \
f"**Temp Updated:** {new_temp}\n" \
f"**Status:** {status}\n" \
f"**__JAE_Processed__**"
# CRM update - only update temp field (custom field on notes)
update_data = {
"title": new_title,
"bodyV2": {"markdown": new_body},
"temp": new_temp
}
data = json.dumps(update_data).encode()
req = urllib.request.Request(
f"{CRM_URL}/notes/{note_id}",
headers={"Authorization": f"Bearer {CRM_TOKEN}", "Content-Type": "application/json"},
data=data, method='PUT'
)
with urllib.request.urlopen(req, timeout=10) as r:
return True, new_title
except Exception as e:
log(f"Update failed: {e}")
return False, title
def process():
s = load_state()
log("=== JAE v3 Starting - Processing ALL notes ===")
notes = fetch_notes()
log(f"Fetched {len(notes)} notes")
for note in notes:
body = note.get('bodyV2', {}).get('markdown', '')
# Skip already processed by v3
if '__JAE_Processed__' in body:
continue
title = note.get('title', '')
note_id = note.get('id')
temp = get_temp(title)
if not temp:
log(f"Skip: no temp in title: {title[:35]}")
continue
url = extract_url(body)
if not url:
log(f"Skip: no URL: {title[:35]}")
continue
log(f"Validating: {url[:45]}")
is_valid, status = validate_website(url)
if is_valid and temp != 'HOT':
new_temp = upgrade(temp)
log(f"UPGRADE: {temp}->{new_temp} | {title[:40]}")
ok, new_title = update_note_full(note_id, body, title, new_temp, status)
if ok:
s['upgraded'] += 1
log(f" Updated title: {new_title[:50]}")
else:
# Still process to set Temp field even if not upgrading
ok, new_title = update_note_full(note_id, body, title, temp, f"verified_{status}")
log(f"Verified: {temp} | {title[:40]}")
s['processed'] += 1
time.sleep(0.5) # Rate limiting
s['last_check'] = datetime.now().isoformat()
save_state(s)
log(f"=== Done: {s['processed']} processed, {s['upgraded']} upgraded ===")
def main():
while True:
process()
log("Waiting 3 hours...")
time.sleep(10800)
if __name__ == "__main__":
main()