BOTWAVE Trades — AI Dispatch for Trade Businesses

── What it looks like

A conversation, not a dashboard.

Crews live in Telegram. The bot lives there too. No new app to install, no login to forget, no training required.

User: who still owes me money

TRADES: Three open invoices.

  Garcia Plumbing Supply    $2,400   NET-30 expires May 12
  M. Torres                 $285     dispatched Apr 28, no reply
  Riverside Apartments LLC  $950     partial — $500 received

Total outstanding: $3,635

Run "/remind Torres" to queue a follow-up message.

─────────────────────────────────────────────

User: what did I make in April

TRADES: April revenue — $8,240 collected across 6 jobs.
Three invoices still pending from that month ($1,235 total).
Net April: $7,005 after materials ($1,205 logged).

─────────────────────────────────────────────

User: schedule Rodriguez for leak repair tomorrow 9am

TRADES: Scheduled.
  Crew: Rodriguez
  Job: Leak repair
  Time: Tue May 12 · 09:00
  Confirmation sent to Rodriguez via Telegram.

🔧

Plumbing

live · beta

⚡

Electrical

in development

❄️

HVAC

in development

🏗️

General Contractor

in development

── Features

What it handles.

Dispatch

SMS and Telegram Dispatch

Jobs assigned by message. Crew members confirm via the same thread. No phone tag, no missed dispatch, no whiteboard.

Inbound job request → assigned crew in under 60 seconds
Crew confirmation tracked in substrate
No-reply after 15 minutes triggers a follow-up automatically

Morning Digest

Daily Crew Digest

Every morning before the first job: open tickets, crew assignments for the day, outstanding invoices. One message, no dashboard.

Sent at configurable time — default 06:30 local
Pulls from substrate — no manual compilation
Replies to digest trigger follow-up actions

Job-Site Check-Ins

Timestamped Arrivals

Crew texts "on site" — arrival logged with timestamp and job ID. Departure same. Time-on-site calculated automatically per job, per crew member.

Check-in / check-out by text — one word
Per-job time logs in substrate
Labor cost per job calculable from logs

Scheduling

Calendar by Message

Schedule a job, reschedule it, check crew availability — by message. The bot resolves conflicts before confirming.

Natural language date parsing — "tomorrow at 9" works
Conflict detection before confirmation
Crew notified automatically on schedule changes

Invoicing

Revenue Tracking by Text

"Who owes me" pulls unpaid invoices. "How much this month" gives collected revenue, pending totals, and materials logged. No spreadsheet required.

Invoice log in SQLite — queryable by message
Payment recorded when you say "Torres paid"
Monthly summary always one message away

Architecture

One Config, Four Trades

The same binary runs all four trades. Set BOTWAVE_TRADE=plumbing and it's a plumbing bot. Electrical, HVAC, GC — same code, different config.

python-telegram-bot + SQLite + JSONL substrate
Local inference — no cloud required
Systemd template unit — one service file, four instances

Trade config — what sets the bot's identity Python

# BOTWAVE_TRADE=plumbing sets everything below via config.json
TRADE_KEY  = os.environ["BOTWAVE_TRADE"].strip().lower()
_TRADE_DIR = (_TRADES / TRADE_KEY).resolve()
_TRADE_CFG = json.loads((_TRADE_DIR / "config.json").read_text())

# Config drives labor rates, company name, bot token, knowledge base.
# Zero hardcoded trade logic in the core binary.
LABOR_RATE = float(os.environ.get(
    "BOTWAVE_LABOR_RATE",
    _TRADE_CFG.get("default_labor_rate", 125)
))

# Four systemd instances from one unit file:
# botwave-sc@plumbing.service   botwave-sc@electrical.service
# botwave-sc@hvac.service       botwave-sc@gc.service

Dispatch by text. Your crew doesn't need an app.

A conversation, not a dashboard.

What it handles.

Dispatch by text.
Your crew doesn't need an app.