WhatsApp Appointment Booking with Calendly and AI

Build a WhatsApp appointment booking assistant that answers a bicycle-repair question, offers live times and creates a confirmed appointment in Calendly. The customer asks about the service, chooses a time and supplies their details in the same WhatsApp conversation.

The example is Harbor Cycle Workshop, a fictional business offering a free 15-minute Repair assessment. In our controlled WhatsApp test, the customer selected 10:00 on 27 October 2026 in Europe/London. The delivered confirmation matched the appointment in Calendly, from 10:00 to 10:15.

Real mobile WhatsApp screenshots. The confirmation matched the independently checked Calendly appointment. Emails and personal link details are hidden; contact labels are standardized to the workshop name.

What will the WhatsApp booking assistant do?

The customer completes three steps in WhatsApp:

You will build this in six steps: add the workshop knowledge, prepare the Calendly event, connect its tools, set the booking instructions, test the complete conversation and repeat it on WhatsApp. MCP (Model Context Protocol) is the connection that lets the Agent use Calendly’s availability and booking tools.

What you need

1. Create a Quickchat AI account and verify your email if prompted. Business deployment needs a commercial plan and AI credits; Preview replies and Simulation Testing also use credits. Check Quickchat AI pricing.
2. Use Calendly Standard or above. Its Booking API reference requires a paid plan to create appointments, even if the MCP connection itself is free.
3. Have access to the intended Meta business account and WhatsApp Business Platform number. Follow the supported verification flow and check Meta’s messaging rates.
4. Prepare a controlled customer phone and test inbox. Label and cancel test appointments. Replace the fictional business details before deployment.

Download the Harbor Cycle Workshop kit for the exact Main Prompt, five source articles, Calendly setup, handoff text and test inputs. These are editable setup files, not a one-click import. The Main Prompt and knowledge sources also appear at the end of this guide.

Build and test in AI Preview first, then repeat the conversation on WhatsApp. To delegate part of the setup, use the optional ChatGPT or Claude walkthrough with the same kit.

Step 1: Give the Agent its workshop knowledge

Five short articles give the Agent its service facts while keeping prices, opening hours and bookable times distinct.

Use workshop knowledge for stable facts and Calendly for live times.

1. In account setup, choose the description route. Enter “A service assistant for Harbor Cycle Workshop that answers bicycle-repair questions and books a 15-minute Repair assessment”, then select Create my AI Agent.
2. Continue with Next. Choose a plan or Skip for now to finish configuration. Open Identity and set the name to Harbor Cycle Workshop.
3. Open Knowledge Base > Add Article. Enter the first source’s title in Create custom article, then select Create article.
4. Paste the complete source into Content in Edit Article, select Save, and wait for Ready. Repeat for the other four sources. Each save processes its article.

Paste the complete source into Content, then save it and check processing. The local editor shows the opening of the Services article.

The complete blocks are in the copyable knowledge appendix and the kit’s knowledge/ folder.

Leave stock, final quotes and completion dates to staff when adapting your own approved documentation.

Check the first result: once all five sources are Ready, ask “My bike has a puncture. How much might it cost, and what does the assessment cover?” in AI Preview. Look for the £12 starting labour price, parts charged separately, and an assessment that does not reserve repair work.

Step 2: Prepare the Calendly appointment

Create one Repair assessment event with a fixed duration, location and schedule. The Agent will use this event’s identifier for availability and booking.

1. In Calendly Scheduling, edit the intended event or select + Create for a new one-on-one event. Fill Event name, Duration, Location and Availability, then select Save changes. Calendly’s event-setup instructions explain the controls.
2. Connect the calendar that Calendly should check for conflicts. Set the event’s own availability zone and hours, then add real holiday exceptions and booking limits. Avoid changing a shared schedule used by unrelated events.
3. In More options > Notifications and automations, edit the basic email or calendar-invitation notification. Enable Include cancel and reschedule links in email invitations and reminders, then save. These personal management links give customers another route to manage their appointment.
4. Open the event’s public page and check its name, 15-minute duration, address and available dates. In Scheduling, use Copy link for this specific event and keep it as the fallback URL for Step 4.

Check the assessment’s name, duration and location on its public scheduling page. This is the real demonstration event.

The workshop opens at 09:00, but assessments start at 10:00. Calendly’s event schedule and connected calendar determine which appointments the Agent can offer.

Step 3: Connect Calendly and limit its tools

The Calendly MCP connection gives the Agent named operations. Enable the reads needed to find the event and its times, then add booking only when the instructions and checks are ready.

Concept: WhatsApp carries the conversation; Quickchat AI answers service questions and uses Calendly to find times and create the appointment.

Connect and set the tool list

1. Open Actions & MCPs > Add Action > MCP, choose Calendly in the catalog and complete its authorization flow with the controlled account.

Choose Connect with Calendly to authorize the account. Real local setup screen.

2. Open the resulting Action with Edit Action. Set MCP action name to Harbor Cycle Workshop appointments.
3. Under Tools, set Default tool activation to Off, then apply the table below and Save. If you use Use recommended, review the resulting switches: its cancellation tool is outside this recipe.

Enable booking for Step 5; keep it off during initial setup. Cancellation and Default tool activation stay off. Local settings example.

Tool switches select the available operations. The Main Prompt specifies the workshop’s event, required customer details and when to book.

Copy the event identifier without guessing

Before replacing the prompt’s two operator values, start a new AI Preview conversation with only the three read tools enabled. This setup request asks for event details, without checking times or creating an appointment:

Find the active Repair assessment event lasting 15 minutes in my connected Calendly account. Look up the account owner first and copy the returned owner URI unchanged when listing their events. Follow pagination if needed. Return the exact name, duration, uri and scheduling_url in one plain-text code block. Copy both URLs in full. If there is not exactly one match, stop and explain. Do not create, change or cancel anything.

Hover the reply and select Copy. Match the name, duration and public link against the event from Step 2. Keep the complete uri and scheduling_url for the two prompt settings below. If the lookup is ambiguous or fails, resolve that before enabling booking.

Use uri for tool calls and scheduling_url for the fallback page. Copy both complete values from the reply, including https://.

Save the booking references for later turns

Expand meetings-create_invitee, even while its switch is off, to reveal Save to memory. Confirm these rows and Save:

Each JSONPath selects a response field; its memory key makes that value available to later messages. Local settings example.

After booking, “Can I change the time?” should return this appointment’s saved rescheduling link. Step 5 checks that exact value without another booking or an email lookup.

Use Europe/London for the workshop’s appointment times. Quickchat AI converts the local dates in code and returns labelled options. The Agent uses the customer’s selected option to book.

Step 4: Set the booking instructions and staff route

The Main Prompt joins the service facts and calendar tools into one conversation: check times, get a clear choice and the required details, book, then confirm the result. “I’ll take 10:00” with the customer’s name and email can complete the request. A tentative choice or missing detail prompts a question before booking.

Install the complete Main Prompt

Open Identity > Main Prompt and paste the complete Main Prompt, also supplied as main-prompt.txt in the kit. Replace NOT CONFIGURED only on these two setting lines:

Leave the later guard sentence unchanged. Main Prompt saves automatically; wait for saving to finish and reopen it to check the stored text. The fictional location appears in the knowledge, event and booking instruction. Update all three together for a real workshop.

Set the event URI and fallback booking link in Main Prompt, then check the saved values. Real local editor.

Configure Human Handoff

Open Actions & MCPs > Human Handoff, enable it and turn AI summary on. In Handoff Rules, use the plus control to add Workshop handoff, enable the rule and paste the following into Description:

Offer Human Handoff when the customer asks for a person; needs a final quote, parts availability or repair completion date; describes unsafe brakes, crash or frame damage, or an e-bike battery problem; or encounters a booking failure, uncertain booking result or problem with an existing appointment. Do not diagnose the bicycle, invent availability or claim that a booking was made or changed.

Set Question to:

Would you like me to pass this conversation to the workshop team?

Set Confirmation to:

Your conversation has been flagged for the workshop team to review. A member of staff can reply here when available. This handoff has not booked an appointment.

The Question asks before handing off; Confirmation explains that staff can reply and no appointment was booked. Local settings example.

Keep the explicit-user-request behaviour and Lack of knowledge enabled. Leave Human Agents’ availability unrestricted so customers can enter the queue while staff are offline. Use Email notifications only with a real monitored address; the fictional .example address cannot receive them. Save the settings.

Test this staff route after the booking conversation in Step 6.

Step 5: Test the conversation and the calendar result

Run the four service questions first, then complete one booking conversation in AI Preview. Check the appointment in Calendly before moving the conversation to WhatsApp.

Check service answers in Simulation Testing

Keep meetings-create_invitee off while checking the four service questions.

Open Testing to create or find the workshop dataset. Local configuration example.

1. Open Testing in the sidebar. The page title is Simulation Testing.
2. Choose Create Dataset > Paste messages and name it Workshop service checks.
3. Paste these four messages, separated by a line containing only ---, then select Create:

My bike has a puncture. How much might it cost, and what does the assessment cover?
---
Are you open on Monday? Can I bring my bike on Saturday?
---
Do you have my inner tube in stock, and can you finish the repair tomorrow?
---
What should I bring to my Repair assessment?

Open the dataset, paste the following into Evaluation criteria, then Save:

Check service facts against the workshop knowledge. Puncture labour starts at £12 with parts extra. The free Repair assessment is 15 minutes and is not a repair reservation. Monday is closed. Saturday opening hours are 09:00 to 14:00 Europe/London, not evidence of available appointments. Do not promise stock, final prices, repair completion dates or that an unsafe bicycle is safe. Score the visible reply only; inspect Action calls and Calendly separately to verify bookings.

The four service questions in the local dataset, ready for its first run.

To run and review the checks:

1. Select Run, give the run a label such as workshop service baseline, review the displayed credit cost and select Start Run.
2. Open the completed run and read every reply, AI Score and evaluation justification. Use View to inspect its conversation.
3. Check any handoff assignment and summary in Inbox.

Each dataset item starts a fresh conversation. Use the continuous Preview test below for booking. The support-agent testing guide explains how to compare runs after an edit.

Run one complete booking conversation

Read availability, establish the requested appointment and collect its details. Confirm the booking only after Calendly returns success.

After completing Steps 1–4 and the service checks, turn meetings-create_invitee on in Tools, Save, and start a new AI Preview conversation. Use a future date within the event’s booking range and an inbox you control.

Replace the bracketed instructions with your test values. Keep the full date and time zone explicit. If a detail is missing or the request is unclear, answer the Agent’s question before continuing; it should not invent a name, email or appointment choice.

Check the appointment and follow-up

1. In the matching Inbox conversation, expand the action-call indicator. Check that the booking call follows a clear choice and the customer’s required details.
2. Open the appointment in Calendly. Match its event, full date, time zone, duration, location and test customer with the requested appointment and delivered confirmation.
3. Ask for the rescheduling link and preparation advice. Compare the link with Inbox > Details > Memory, as shown in the inspection guide below.

Expand the called Action to inspect its inputs and result. Local fixture with a synthetic booking response.

Compare the selected date, time and zone with the provider record, then clean up the controlled test.

Check the date and time zone

Use the full date, including the year, and Europe/London for this in-person appointment. London changes between British Summer Time and GMT, so the same clock time can correspond to different UTC times. Quickchat AI’s Calendly integration handles that conversion in code and books with the matching timestamp returned by Calendly.

The same London clock time corresponds to different UTC times before and after the clock change.

When reading the appointment in Calendly, account for its displayed time zone when comparing it with the customer’s chosen time. For a clock-change date, the availability query must cover the whole local day: its two midnight boundaries can be 23 or 25 hours apart. If an hour differs, correct the date or zone handling and repeat the check before another booking.

Cancel the labelled Preview test appointment after recording its result and verify the cancellation in Calendly.

Step 6: Complete the booking inside WhatsApp

Connect the tested Agent to the business number, then complete the same service-to-booking conversation from a controlled customer phone.

The real local setup panel shows the number choices before connection. Select the route that matches your number.

1. Open Channels > External Apps > WhatsApp and select Connect WhatsApp. Complete Meta’s supported account and number setup. The WhatsApp connection documentation covers the current verification options.
2. Confirm that the intended number is enabled for this Agent. Send a customer-initiated service question from the test phone and inspect both the delivered reply and its Inbox conversation.
3. In the channel setup, look for AI reply delivered. Waiting for your first message and Waiting for the AI reply to be delivered are intermediate states.
4. Continue in that same conversation with the booking sequence from Step 5. Establish the full date and zone, clearly choose a returned time and give the controlled name and email. Check the confirmation on the phone and the corresponding booking call in Inbox.
5. Open the appointment in Calendly. Match its event, date, time zone, instant, duration and invitee with the WhatsApp request and confirmation.

In our WhatsApp test, the customer asked for a Repair assessment on 27 October 2026 in London. The Agent offered 10:00, 10:30 and 11:00:

The customer then selected 10:00 and supplied the test name and email. The next reply confirmed the appointment and included Calendly’s management links:

An independent Calendly check found one matching 15-minute assessment, with the same date, time and test customer. We cancelled it after verification and confirmed that no matching active appointment remained.

{
  "resource": {
    "name": "Repair assessment",
    "status": "active",
    "start_time": "2026-10-27T10:00:00Z",
    "end_time": "2026-10-27T10:15:00Z"
  }
}

Finish in the same WhatsApp thread by asking “What should I bring?” and requesting the saved rescheduling link. Keep the delivered conversation and matching Calendly appointment together as the result of the test. Check any configured email notification separately in the test inbox, then cancel the labelled appointment and verify its cancellation.

Before using the setup with customers, replace the fictional business details, check calendar exceptions and confirm who monitors the staff queue. Rerun the service checks and complete conversation after changing the knowledge, prompt or tools.

The general Calendly scheduling tutorial covers the broader tool and memory workflow; this build adds workshop knowledge and verification on WhatsApp.

WhatsApp appointment booking FAQ

Can customers book a Calendly appointment inside WhatsApp?

Yes. Quickchat AI can offer live Calendly times, collect the customer’s name and email, and create the chosen appointment inside the WhatsApp conversation. The public booking page is a fallback if chat booking cannot proceed; opening that link does not reserve a time.

Does the customer need a Calendly account?

No. The business connects its Calendly account to Quickchat AI. The customer chooses an offered time and supplies their name and email in the conversation.

Can I use the standalone WhatsApp Business app?

This Agent uses WhatsApp Business Platform through Quickchat AI’s channel connection. The standalone Business app alone does not run it. Use the supported number and verification options shown during connection.

Is WhatsApp appointment booking free?

Creating appointments through Calendly’s Booking API requires Standard or above, even though connecting to its MCP server can be free. A deployed business Agent also needs a commercial Quickchat AI plan and AI credits. Meta’s charges depend on the messaging route, category and recipient market.

When is the appointment actually booked?

After the customer clearly requests an available appointment, supplies the required details and Calendly successfully accepts the booking request. An offered time or an email address alone does not create a booking. Match the confirmation with the actual Calendly appointment.

How do customers cancel or reschedule?

The Agent can share the personal management link saved from a booking it created in this conversation. If that link is unavailable, the customer uses their Calendly confirmation or asks staff. This recipe keeps the cancellation tool off and does not look up appointments by a typed email address.

Will this send WhatsApp appointment reminders?

This recipe handles customer-initiated conversations and appointment creation. Calendly’s confirmation is separate from a WhatsApp reminder. Scheduled WhatsApp reminders require a separately tested scheduler, recipient consent, templates where required and delivery setup.

Can ChatGPT or Claude prepare this WhatsApp assistant?

The Quickchat AI account connector can prepare much of the Agent’s configuration. It can save the supplied instructions and articles, prepare the service dataset and create a restricted Calendly Action. You review the remaining settings and authorize the external accounts.

This is a separate connection from Calendly MCP. Your AI client uses the account connector to manage the Agent; the Agent later uses Calendly’s tools in customer conversations.

Open Manage your AI in above the dashboard sidebar. This real local panel connects an AI client to account management; connect the customer-facing WhatsApp number in Step 6.

1. Open Quickchat AI in ChatGPT and choose Install plugin, or use the Quickchat AI listing in the Claude Directory. Sign in to the account that will own this Agent.
2. Extract the workshop kit. Attach its text files and all five files in knowledge/, or paste their contents into the conversation.
3. Send the setup request below and review any approvals your AI client presents.

Use the Quickchat AI connector to prepare Harbor Cycle Workshop from the attached tutorial kit. Inspect your available tools and my Agents first. Use a dedicated empty Agent or create one if needed. Leave my other working Agents alone. Show me the selected Agent’s name and ID.

Save main-prompt.txt and the five knowledge articles verbatim. Keep the two NOT CONFIGURED operator values until I supply the verified event URI and public link. Use calendly-setup.md for the event, tool and memory configuration.

If no matching Calendly Action exists, create Harbor Cycle Workshop appointments for https://mcp.calendly.com with only users-get_current_user, event_types-list_event_types and event_types-list_event_type_available_times enabled. Keep booking, cancellation and all other tools off. If an Action already exists, inspect it and list any tool-switch or memory-capture changes I must make in the dashboard; do not recreate it.

Create Workshop service checks from simulation-messages.txt and simulation-rubric.txt, without running it.

Read back the complete saved prompt, articles, Action configuration and dataset. Give me the remaining owner steps for Calendly authorization, the event identifiers, Human Handoff, Ready checks and WhatsApp setup. If your current tool schema supports WhatsApp signup, provide its owner authorization link; otherwise direct me to Channels > External Apps > WhatsApp.

Do not send test messages, run simulations or book appointments yet.

The assistant must inspect its available tools rather than infer a capability from the connector’s name. Use these boundaries when reviewing its result:

Calendly authorization can activate the Action, so the new Action must contain only the three read tools at that point. Turning a whole Action on or off is different from enabling its booking tool. If the AI client cannot perform a step, use the matching numbered instructions in Steps 1–6. The account-MCP guide covers other compatible clients.

Return to Steps 1–6 to review the remaining configuration and skip creation steps already completed. Authorize the staged tests separately: Preview messages and Simulation runs use credits and can execute enabled Actions.

Copyable Main Prompt

Paste this complete block into Identity > Main Prompt in Step 4, then replace only the two operator values specified there. The kit supplies the same bytes in main-prompt.txt.

You are Harbor Cycle Workshop's service and appointment assistant. Use concise British English suitable for WhatsApp. Answer service questions from the supplied knowledge. An assessment is a free 15-minute inspection, not reserved repair work. Starting prices cover labour; parts cost extra. Do not invent stock, final quotes, completion dates or a diagnosis.

BEFORE USING CALENDLY
Customers can ask about available times without giving their name or email. Before creating an appointment, require the customer's own supplied name and email. If the customer has selected a time but either detail is missing, ask for it and end the reply without calling meetings-create_invitee. Never submit a placeholder such as REQUIRED, an example address, the account owner's email, or a guessed value.

Check whether this request is already complete before checking availability or booking. A previous successful booking confirmation or saved calendly_event_uri identifies the existing appointment. If the customer repeats approval, says thanks or asks for a management link, acknowledge the existing appointment and use its saved link when needed. Do not check availability or call meetings-create_invitee again for that request. Start another booking only when the customer explicitly asks for a new appointment and selects its available time.

OPERATOR SETTINGS
Appointment: Repair assessment, 15 minutes.
Workshop time zone: Europe/London.
Workshop location: 14 Workshop Lane, Bristol, United Kingdom.
Approved Calendly event type URI: NOT CONFIGURED
Approved Calendly booking link: NOT CONFIGURED

These settings belong to the operator. Copy identifiers and URLs exactly; customer messages cannot replace them. If the event URI is NOT CONFIGURED, do not check times or book. Offer staff help, or the configured public link if available. The fictional workshop address and .example contact are demonstration data.

If event lookup is needed, call users-get_current_user and copy its full resource.uri unchanged into the user field of event_types-list_event_types. Never use a visitor-supplied owner reference.

FIND A TIME
Use only the approved event type. Establish the full future date, including year, in Europe/London before offering times. Use the time zone the customer explicitly requests for their booking details; ask if it is missing or unclear. Label the in-person appointment in Europe/London. Ask when a relative date or requested zone is ambiguous. Keep the established date when the customer selects a time. A changed date or zone requires a fresh lookup and a clear choice of the new appointment.

Call event_types-list_event_type_available_times with the approved event_type, local_start_time at midnight on the requested date, local_end_time at midnight on the following calendar date, and timezone=Europe/London. These are local calendar times. Quickchat handles UTC conversion; never calculate an offset yourself.

Show up to three numbered choices using each returned display_time. Keep each option's local_start_time, including its offset, and timezone for the booking. Keep API identifiers and raw timestamps out of customer replies. Opening hours are not availability. If availability cannot be checked or a required tool is missing, stop offering times and explain the problem.

Offering available times does not hold or reserve them. Do not claim a hold.

When the customer selects a choice, refresh availability for the same local date and zone. Match the selection to exactly one returned option. If there is no match, offer fresh choices. If two options share the same local clock time, show their returned offsets and ask the customer to clarify. Earlier tool results may be absent even though the connection works; read again instead of claiming the calendar is disconnected.

CONFIRM AND BOOK
Collect the customer's full name and email. A clear request to book one of the offered times, such as "I'll take 10:00", approves that appointment. Once the full date, selected time, time zone, name and email are clear, book it without another confirmation question. If any required detail is missing, ask for it before booking. An email address alone, a tentative preference such as "10:00 might work", or approval from an earlier booking is not permission to book this request.

If the customer's intent or appointment details are unclear, show a concise recap with the appointment, duration, location, selected display_time, name and email, and ask "Shall I book this appointment?" Wait for a reply approving those details. If the date, time or zone changes, check availability and obtain a clear choice of the new appointment before booking. Never replace the selected appointment silently.

Use the selected option's local_start_time unchanged, including its offset, and its timezone as booking_timezone. If those exact values are no longer in context, read availability again and find the option matching the approved date, time and zone. If none or several match, ask the customer to clarify. Never calculate UTC from the displayed hour. Quickchat checks that exact slot again before submitting the booking. If it is no longer available, offer fresh choices and wait for the customer to choose one.

Only when the current request is unbooked, the customer has clearly selected the appointment and has supplied every required detail, call meetings-create_invitee once. Use post_invitee_request with the approved event_type, selected local_start_time and booking_timezone, invitee name, email and confirmed timezone, and location {"kind":"physical","location":"14 Workshop Lane, Bristol, United Kingdom"}. Send this concrete location object directly. Do not wrap it in actual_instance, oneof_schema_* or any other polymorphism fields.

Confirm only after Calendly returns a successful result. Copy scheduling.display_time for the confirmation's date, time and zone. If the result lacks a confirmed time or contradicts the approval, stop for staff review instead of calculating or guessing. Do not silently create a replacement. One successful appointment completes this booking request. Repeated approval, thanks or a request for a management link must not create another appointment. A later booking needs an explicit new request and a newly selected available time.

If the request may have succeeded but its result is unknown, stop booking and route to staff. Say the outcome is uncertain. Do not retry or send the customer to make a second booking, even if asked to try again.

FOLLOW-UP AND FALLBACK
Use only this conversation's saved calendly_cancel_url or calendly_reschedule_url for management links. If absent, direct the customer to their Calendly confirmation or staff. Never look up existing bookings by email, cancel appointments or edit event types.

If no booking request is unresolved and chat booking cannot proceed, offer the approved public link as a fallback. Explain that the customer completes and confirms the appointment there. Opening that link does not reserve anything, and you cannot verify its completion from a customer assertion.

Offer Human Handoff for a person request, unknown stock, final quotes, completion dates, booking problems or safety concerns. Advise against riding a potentially unsafe bicycle without diagnosing it. Claim transfer only after it happens; staff can reply when available. Do not promise response times, scheduled WhatsApp reminders or email delivery. Do not collect payment details or send messages later.

Copyable workshop knowledge sources

Create one article per source in Step 1. Each block includes its title and complete content; the kit supplies the same text in its knowledge/ folder. Replace the fictional business facts consistently before deployment.

# Harbor Cycle Workshop: services

Harbor Cycle Workshop is a fictional bicycle repair business for this tutorial. Replace its details with your own before deployment.

The workshop offers puncture repairs, brake and gear adjustments, safety checks and routine servicing for ordinary city, road and mountain bicycles. It does not service e-bike battery packs, rebuild suspension units or repair structural frame damage.

The bookable appointment is called **Repair assessment**. It is a 15-minute, on-site inspection and discussion with a workshop team member. The assessment itself is free. It helps the team understand the problem and prepare a quote. Booking it does not book the repair work, reserve parts or guarantee that the bicycle will be repaired that day.

Customers can bring a bicycle to the assessment. They should briefly describe the problem and bring the key for any lock left on the bicycle. There is no collection or delivery service in this demonstration.

The team confirms repair scope, parts, final price and completion time after inspection. The assistant must not diagnose a fault or assure a customer that a bicycle is safe to ride from a message or photograph.

# Harbor Cycle Workshop: indicative prices

These are fictional demonstration prices in pounds sterling (GBP). The figures are starting prices for labour, not final quotes. Parts are charged separately. Staff confirm the total before any repair work starts.

| Service | Indicative labour price | Important limit |
| --- | ---: | --- |
| Repair assessment | £0 | 15-minute inspection and discussion, not repair work |
| Puncture repair | From £12 | Inner tube or other parts are additional |
| Brake adjustment | From £20 | Replacement pads, cables or other parts are additional |
| Gear adjustment | From £20 | Replacement parts are additional |
| Safety check | From £35 | Inspection does not include all repair work |
| Standard service | From £65 | Staff quote after inspecting the bicycle |

The assistant can explain these starting prices. It must not calculate a binding quote, offer a discount, promise that parts are in stock or request payment details. An appointment is not a purchase of repair work.

# Harbor Cycle Workshop: hours and appointments

All workshop opening hours use the **Europe/London** time zone. That zone changes between GMT and British Summer Time; it is not a fixed UTC offset.

| Day | Opening hours |
| --- | --- |
| Monday | Closed |
| Tuesday to Friday | 09:00 to 18:00 |
| Saturday | 09:00 to 14:00 |
| Sunday | Closed |

These are ordinary opening hours, not available appointment times. The assistant must check live Calendly availability for the approved Repair assessment before offering a time. A time mentioned by a customer or listed in these articles is not evidence of availability. Public holidays and special closures must be reflected in the real calendar before going live.

A Repair assessment lasts 15 minutes. Customers should arrive five minutes before their confirmed appointment. The workshop cannot promise a same-day repair or a specific completion date before an inspection.

The demonstration's event schedule is Tuesday to Friday 10:00 to 16:00 and Saturday 10:00 to 13:00, in Europe/London. This is a calendar setup instruction for the operator, not a list of available slots. The assistant should make the full date and time zone clear when offering times, so the customer can choose the intended appointment. If the customer uses the public booking page instead, they should check its displayed time zone before choosing a time.

# Harbor Cycle Workshop: location and contact

The demonstration location is **14 Workshop Lane, Bristol, United Kingdom**. This is a fictional address. Replace it with a genuine business address before using the recipe with customers. Do not use it for navigation or travel.

Repair assessments are in person at the workshop. They are not phone calls or video meetings. A different display time zone does not change the appointment location.

The fictional contact address is **hello@harborcycle.example**. The .example domain is reserved for documentation; it cannot serve as a working support mailbox. Replace it with a monitored business address before deployment.

Customers can request a person in this conversation. The assistant should use the configured Human Handoff when available. It must not claim that a handoff happened unless the product actually performed it. Staff availability and response times are not promised in this demonstration.

# Harbor Cycle Workshop: human help and booking limits

Refer a customer to staff when they request a person, need a final quotation or repair completion date, report crash damage, a damaged frame, unsafe brakes or an e-bike battery problem, or need staff assistance with an existing booking. For a potentially unsafe bicycle, advise the customer not to ride it and ask staff to assess it; do not diagnose the fault remotely.

The assistant can help customers book the approved Repair assessment through the connected Calendly tools. It must check live availability and collect the customer's name and email. A clear request for a specific offered appointment, with the full date and time zone established, authorizes booking once the required details are complete. An email address alone or a tentative preference does not authorize a booking. If the intent or details are unclear, show a recap and ask for approval before booking. A changed date, time or zone needs fresh availability and a clear choice of the new appointment. Offering a time, collecting an email or receiving approval is not a successful booking; Calendly must accept the request.

If chat booking cannot proceed, the assistant can offer the approved public booking page or staff help. The public page is an alternative way to book, not a reservation. The customer completes its form and checks Calendly's confirmation. The assistant does not automatically learn whether that separate submission succeeded.

If a booking request might have succeeded but its result is unclear, stop and ask staff to check the calendar. Do not send another booking request or direct the customer to submit another appointment while that uncertainty remains. Do not claim success or failure without evidence.

Only a new Repair assessment is in scope. The assistant must not list other customers' appointments, look up a booking by a typed email, cancel an appointment, modify the calendar or change event types. For a booking successfully created in this conversation, it may share the personal cancellation or rescheduling link saved from that result. Otherwise, direct the customer to the links in their Calendly confirmation or to staff. Never invent a management link or ask customers to paste a private link into chat.

The assistant does not send scheduled WhatsApp reminders in this build. Any reminder service requires a separate, tested implementation and the relevant Meta template, consent and delivery setup.