Free OpenClaw Skill for Pizza Restaurants: Delivery and Dough Cycle Agent (Download)

What this skill does for your pizza restaurants

• Routes inbound orders across Toast online ordering, Slice, ChowNow, and Olo with the per-channel commission math factored into the pricing decision.
• Dispatches delivery across in-house drivers, DoorDash Drive, and Onfleet with the per-mile rate and the expected wait time per channel.
• Schedules the dough cold-ferment on a 24, 48, and 72-hour cycle, with the day's batch size keyed to historical pie velocity by day-of-week.
• Audits the Restaurant365 prime-cost report and flags weekly food and labor variance against the operator's target band.
• Runs the Friday and Saturday peak-hour pre-prep plan so the line is staged with parbaked dough, cheese pulled, and toppings ready before the rush.

What's inside this OpenClaw skill

The full file you download, indexed below for search engines and for operators who want to read before they install.

---
name: pizza-delivery-and-dough-cycle
description: Delivery dispatch, dough cold-ferment scheduling, marketplace routing, and Restaurant365 prime-cost watch for pizza restaurants. Integrates with Toast, Slice, ChowNow, Olo, DoorDash Drive, Onfleet.
version: 1.0.0
author: OpenClaw Consult (Adhiraj Hangal)
license: MIT
url: https://openclawconsult.com/skills/pizza-restaurants
---

# OpenClaw Skill: Pizza Restaurant Delivery and Dough Cycle Agent

## Overview

This skill turns the OpenClaw agent into the dispatch, dough-cycle, and
marketplace coordination layer for a pizza restaurant. It owns delivery
routing across in-house drivers, DoorDash Drive, and Onfleet; dough cold-
ferment scheduling on a 24-72h cycle; marketplace routing across Slice,
ChowNow, Olo, and Toast online; and the Restaurant365 prime-cost watch.

It is designed for independent pizza restaurants and small chains (1 to 5
shops), running 150 to 600 orders per day, on Toast for the POS.

## What this skill does

1. Routes inbound orders across Toast, Slice, ChowNow, Olo with commission
   math
2. Dispatches delivery across in-house drivers, DoorDash Drive, Onfleet
3. Schedules dough cold-ferment on 24, 48, 72h cycle
4. Audits R365 prime-cost weekly and flags variance from target
5. Runs Friday and Saturday peak-hour pre-prep plan
6. Routes orders over $200, catering orders, and dough variance over 10% to
   a human

## Triggers

```yaml
triggers:
  - type: heartbeat
    schedule: "0 14 * * *"          # Daily 2pm
    action: plan_tonight_dough_batch_for_two_days_out

  - type: heartbeat
    schedule: "0 9 * * 1"           # Mondays 9am
    action: review_prime_cost_last_week

  - type: heartbeat
    schedule: "0 15 * * 4"          # Thursdays 3pm
    action: plan_friday_peak_pre_prep

  - type: on_event
    event: order.created
    action: route_to_marketplace_and_assign_delivery

  - type: on_event
    event: dispatch.driver_returning
    action: assign_next_order_in_queue

  - type: on_event
    event: dough.batch_used_up
    action: log_actual_vs_planned_for_refinement
```

## Workflow: marketplace order routing

When a new order comes in, the skill:

1. Identifies the channel: Toast direct, Slice, ChowNow, Olo, or phone
2. Applies the per-channel commission math:
   - Slice marketplace: 10-15% commission typical
   - Slice register-direct: lower
   - ChowNow: flat monthly + small per-order
   - Olo: contract-specific
   - Toast direct: payment processing only
3. Calculates the net per-pie margin
4. Routes to the kitchen ticket printer
5. If the order is over the autonomous threshold (default $200), routes to
   the manager for sign-off

The skill surfaces per-channel margin in the weekly digest so the operator
can see which marketplace is paying.

## Workflow: delivery dispatch

When an order is ready for delivery, the skill:

1. Checks in-house driver availability and current driver locations
2. Calculates expected delivery time for in-house dispatch
3. If in-house wait would exceed the customer's promised window, checks
   DoorDash Drive availability and rate
4. Compares in-house cost vs DoorDash Drive cost
5. Routes to the cheaper option that meets the time window
6. Dispatches via Onfleet for in-house or via DoorDash Drive API for on-
   demand
7. Texts the customer the driver's name and ETA

The skill respects the manager's policy on autonomous DoorDash Drive
dispatch (cost ceiling, time-window threshold). On rules edges, it routes
to the manager.

## Workflow: dough cold-ferment scheduling

The skill maintains a per-day dough target:

- Monday: low (post-weekend recovery)
- Tuesday-Thursday: midweek baseline
- Friday: high (peak)
- Saturday: high (peak)
- Sunday: mid (steady)

The target is keyed to historical pie velocity by day-of-week and weather.
For a 48-72h cold ferment, the skill plans the mix two-to-three nights
ahead of service. Each evening at 2pm, the skill:

1. Confirms the next-night-mix target
2. Adjusts up or down based on the next 2-3 days' weather forecast
3. Posts the dough plan to the kitchen
4. Logs the actual usage at end-of-night to refine the target over time

The skill tracks variance: planned vs actual. After 4 weeks it can predict
within 5-8% of actual usage.

## Workflow: Friday peak pre-prep

Every Thursday at 3pm the agent drafts the Friday peak pre-prep plan:

- Parbaked personal-size shells for the early rush
- Mozzarella to pull at 4pm (so cheese is at room temp by 5pm rush)
- Pepperoni and sausage portioned by 3pm
- Garlic-knot dough portioned during the afternoon lull
- Salad prep done in the slow hours
- Soda and beer stocking
- Box assembly count for the projected ticket volume

The plan goes to the kitchen manager. The Saturday version runs Friday at
3pm with the same pattern.

## Workflow: R365 prime-cost watch

Each Monday morning the skill ingests last week's Restaurant365 export and
calculates:

- Food cost as percentage of sales
- Labor cost as percentage of sales
- Prime cost (food + labor) as percentage of sales
- Variance vs target band (typical pizza target: 58-62%)
- Top variance driver: cheese cost spike, overtime, low sales day, etc.

The flag goes to the operator with a one-line attribution. The skill does
not auto-fix prime-cost issues; it surfaces the variance and the likely
cause.

## Memory keys

```yaml
memory:
  - key: pizza.dough_plan[{date}]
    description: Per-day dough target and actual
    schema: { planned: int, actual: int, variance_pct: number, day_of_week: string, weather: string }

  - key: pizza.marketplace_margin[{channel}]
    description: Per-channel net margin rolling average
    schema: { channel: string, commission_pct: number, orders_30d: int, gross_30d: number, net_30d: number }

  - key: pizza.delivery_dispatch_log[{order_id}]
    description: Delivery routing decision
    schema: { provider: enum, cost: number, customer_window: int, actual_delivery_time: int }

  - key: pizza.prime_cost_weekly[{week_starting}]
    description: Weekly prime-cost snapshot
    schema: { food_cost_pct: number, labor_cost_pct: number, prime_cost_pct: number, target_band: string, variance_driver: string }

  - key: pizza.peak_prep_plan[{date}]
    description: Pre-prep plan for a peak shift
    schema: { parbaked_shells: int, cheese_pull_time: time, sausage_portion_count: int, planned_ticket_count: int }
```

## Message templates

```yaml
templates:
  customer_delivery_eta:
    channel: sms
    body: |
      Hi {customer_first_name}, this is {pizza_shop_name}. Your order
      ({order_short_summary}) is on the way with {driver_first_name}. ETA
      {eta_minutes} minutes. Track here: {tracking_url}

  dough_plan_kitchen:
    channel: internal_doc
    body: |
      Dough plan for {service_day} service ({day_of_week}):

      Mix tonight: {planned_dough_balls} dough balls
      Cold ferment window: {ferment_hours}h (until {service_day} at
      {service_window_start})
      Hydration: {hydration_pct}%
      Salt: {salt_pct}%
      Yeast: {yeast_pct}%

      Last {day_of_week} actual: {last_same_day_actual}
      Weather adjustment: {weather_adjustment_summary}
      Confidence: {confidence_level}

  friday_peak_pre_prep:
    channel: internal_doc
    body: |
      FRIDAY PEAK PRE-PREP for {friday_date}:

      Projected ticket count: {projected_tickets}
      Projected gross: {projected_gross}

      Pre-prep tasks (in order):
      1. Parbake {parbaked_shells_count} personal-size shells (by 3pm)
      2. Portion sausage and pepperoni (by 3pm)
      3. Portion garlic-knot dough (3-4pm during lull)
      4. Pull mozzarella from walk-in (4pm sharp)
      5. Box assembly: build {box_count} ready boxes (by 4:30pm)
      6. Salad station prep (by 4pm)
      7. Soda and beer stock check (by 4pm)

      Staffing tonight: {scheduled_staff_summary}

  weekly_prime_cost_alert_in_band:
    channel: sms
    body: |
      Last week prime cost: {prime_cost_pct}% (target band {target_low}%
      to {target_high}%). In band. Food: {food_cost_pct}%, Labor:
      {labor_cost_pct}%. Top variance driver: {top_variance_driver}.

  weekly_prime_cost_alert_out_of_band:
    channel: sms
    body: |
      Last week prime cost: {prime_cost_pct}% (target band {target_low}%
      to {target_high}%). OUT OF BAND.

      Food: {food_cost_pct}%, Labor: {labor_cost_pct}%.
      Top variance driver: {top_variance_driver}.
      Suggested action: {suggested_action}.
```

## Required integrations

| Integration            | Purpose                                       | Read | Write    |
|------------------------|-----------------------------------------------|------|----------|
| Toast                  | POS sales, orders, online ordering            | yes  | optional |
| Slice                  | Marketplace orders                            | yes  | yes      |
| ChowNow                | Marketplace orders                            | yes  | yes      |
| Olo                    | Marketplace orders                            | yes  | yes      |
| DoorDash Drive         | On-demand delivery dispatch                   | yes  | yes      |
| Onfleet                | In-house driver dispatch                      | yes  | yes      |
| Restaurant365          | Prime-cost data                               | yes  | no       |
| Twilio (or similar)    | Outbound SMS                                  | no   | yes      |
| OpenClaw Memory        | Cross-run state                               | yes  | yes      |
| OpenClaw Heartbeat     | Scheduled cadence triggers                    | yes  | yes      |

DoorDash Drive and Onfleet are write integrations because dispatch is the
core action. Slice and ChowNow are typically read for orders and write for
order status updates back to the marketplace.

## Operations safety notes

The skill will not auto-route an order through DoorDash Drive without
respecting the manager's cost ceiling. Default behavior caps DoorDash Drive
dispatch at 1.5x the in-house equivalent cost; orders above that route to
the manager.

Dough variance over 10% from target flags to the operator. The skill does
not auto-adjust the next batch by more than 10% without a manager
confirmation; small adjustments compound into large mistakes without human
context.

## Configuration

```yaml
config:
  shop_name: "Your Pizza Shop"
  approval_mode: manager_approves_all   # or "autonomous_after_2_weeks"
  marketplace_commission_rates:
    slice_marketplace: 0.13
    slice_register: 0.025
    chownow: 0.04
    olo: 0.06
    toast_direct: 0.0
  dough_recipe:
    hydration_pct: 65
    salt_pct: 2.0
    yeast_pct: 0.3
    ferment_hours: 48
  dough_targets_by_day:
    monday: 32
    tuesday: 40
    wednesday: 44
    thursday: 48
    friday: 72
    saturday: 80
    sunday: 56
  delivery_providers:
    in_house:
      cost_per_delivery: 6
    doordash_drive:
      base_fee: 8
      per_mile: 0.85
    onfleet:
      cost_per_delivery: 5
  prime_cost_target_band:
    low: 0.58
    high: 0.62
  autonomous_order_threshold: 200
```

## Quick install

```bash
mkdir -p ~/.claude/skills
mv pizza-restaurants.md ~/.claude/skills/pizza-delivery-and-dough-cycle.md
# Edit the config block to match your shop
```

Launch OpenClaw and the skill is live. Run `plan tonight dough batch`,
`review prime cost last week`, or `plan friday peak pre prep` to invoke it
on demand, or wait for Heartbeat.

## Want it wired to your stack?

This SKILL.md is the starting point. If you want it plugged into your specific
POS, marketplaces, delivery providers, dough cycle, and prime-cost target on
a fixed-price two-week clock, with the founder writing every line of the
build, the door is at https://www.adhirajhangal.com/work

Install in 60 seconds

1. 1

   Drop the file into your OpenClaw skills directory

   Save the downloaded pizza-restaurants.md as ~/.claude/skills/pizza-delivery-and-dough-cycle.md (or scope it in your project-level .claude/skills/). OpenClaw picks up new skill files on next agent launch.

2. 2

   Connect Toast, Slice, ChowNow, Olo, DoorDash Drive, and Onfleet

   Edit the integrations block to point at Toast for the POS, Slice / ChowNow / Olo for the marketplaces, DoorDash Drive for the on-demand delivery API, and Onfleet for in-house driver dispatch. Read access to orders, tickets, and dispatch is enough to start.

3. 3

   Launch in approval mode, then move to autonomous on safe templates

   Start with the manager approving every dispatch decision and dough batch plan for two weeks. The skill's defaults route any order over $200, every catering pizza order, and any dough variance over 10% from target to a human. After two weeks, flip standard delivery dispatch and daily dough cycles to autonomous.

Example conversation

Common questions

Related free skills