← All cookbooks Customer Support cookbook

Customer Support

Ticket triage in real time. Macro suggestions that match your voice. Escalation routing that respects who's on-call.

Marquee recipe

Real-time ticket triage with macro suggestion

First-response time drops; quality stays high.

Trigger
Routine, API trigger fired by Intercom new-conversation webhook
Skills
ticket-triage, draft-response, customer-escalation
Features
Routines (API trigger), Intercom + HubSpot connectors, plan mode for the draft, Slack escalation routing

Every new ticket arrives with a pre-classified urgency, account context from HubSpot, and a draft response in the right tone for the customer segment. The agent reads the draft, makes a tweak, sends. First-response time drops; quality stays high.

How to set it up

  1. 1
    Install the customer-support plugin
    Install in Cowork ↗ claude plugin install customer-support@pace
  2. 2
    Authorize Intercom, HubSpot, Slack

    Intercom for the ticket queue; HubSpot for account context; Slack for escalations.

  3. 3
    Create an API-triggered routine

    In Cowork, open RoutinesNew routineAPI trigger. Copy the URL + bearer token after save.

  4. 4
    Wire Intercom to POST new conversations

    In Intercom: Settings → Webhooks → New webhook, subscribe to "conversation.user.created", point at the routine URL with the bearer token. Then paste this routine prompt:

    Routine prompt
    A new support ticket arrived. The conversation body is
    in the incoming text field.
    
      1. Run ticket-triage to classify
         urgency (P1/P2/P3) and category.
      2. Pull the customer's account context from HubSpot
         (plan tier, recent activity, prior tickets).
      3. Run draft-response : tone
         should match the segment (enterprise / SMB / free).
         Park the draft in the conversation as a private note
         for the human agent to review before sending.
      4. If P1 OR enterprise tier: run
         customer-escalation to ping the
         on-call engineer in #support-escalations.
  5. 5
    Send a test ticket to verify

    Open a test conversation in Intercom from a sandbox account. Within ~30 seconds you should see a private note with the draft response. Once the format is right, every new ticket gets the same treatment.

Saturday on-call digest, no human babysitting

Trigger
Routine, schedule, weekends 09:00 + 18:00
Skills
triage, escalate
Features
Routines (cloud schedule), Intercom + PagerDuty connectors, conditional escalation

Twice on weekends, the routine sweeps Intercom. Routine tickets get auto-responses where the answer is in the knowledge base. The two or three that look real get summarized and paged to the on-call engineer with full context. The Sunday-night Slack thread no longer scrolls.

How to set it up

  1. 1
    Create the scheduled routine in Cowork

    In Cowork, open RoutinesNew routineSchedule. Match the cadence described in the trigger above.

  2. 2
    Paste this as the routine prompt
    Weekend on-call prompt
    Twice on weekends, at 9am and 6pm, sweep the Intercom inbox. For routine tickets where the answer is in the knowledge base, draft a response in our voice and post it. For the two or three that look real (production issue, customer at risk, escalation request), summarize with full context (account, history, what they tried) and page the on-call engineer via PagerDuty.
    
    Post a digest of what was handled vs. what was paged to #support-weekend.
  3. 3
    Click Run now to verify

    Eyeball the first run for accuracy. Once it looks right, the routine fires on its trigger from then on. Assumes the plugin from the marquee above is already installed and connectors authorized.

Turn five resolved tickets into one KB article, every week

Trigger
Routine, schedule, weekly Friday 16:00
Skills
knowledge-base
Features
Routines (cloud schedule), Intercom + Notion connectors, sub-agent per theme

Reads the week's resolved tickets, clusters them by theme, and for each theme with three or more occurrences, drafts a knowledge-base article. Plan mode stages the drafts in Notion; you publish the ones worth publishing. The KB grows from real volume instead of someone deciding what should be there in the abstract.

How to set it up

  1. 1
    Create the scheduled routine in Cowork

    In Cowork, open RoutinesNew routineSchedule. Match the cadence described in the trigger above.

  2. 2
    Paste this as the routine prompt
    KB roll-up prompt
    Every Friday at 4pm, read this week's resolved tickets. Cluster them by theme (what the customer was trying to do). For each theme with 3+ occurrences, draft a knowledge-base article: title, problem, solution, common gotchas, related links.
    
    Stage the drafts in Notion under #support-kb-drafts. We'll publish the ones worth publishing in next week's KB review.
  3. 3
    Click Run now to verify

    Eyeball the first run for accuracy. Once it looks right, the routine fires on its trigger from then on. Assumes the plugin from the marquee above is already installed and connectors authorized.

Customer context on every escalation, before engineering picks it up

When support escalates a ticket to engineering, a routine assembles the customer context first: account size, recent activity, open tickets, deal stage if any, prior escalations and their outcomes. Engineering doesn't ask support "is this customer important?"; the answer is already in the escalation.

Trigger
Routine, Intercom webhook on ticket-escalated
Skills
customer-escalation, customer-context
Features
Routines (webhook), HubSpot + Intercom + Slack connectors

How to set it up

  1. 1
    Create the API-triggered routine

    In Cowork, open RoutinesNew routineAPI. Save and copy the routine URL + bearer token; point the upstream webhook at it (e.g. calendar provider, Intercom webhook, PagerDuty webhook, depending on the trigger).

  2. 2
    Paste this as the routine prompt
    Escalation-context prompt
    When support escalates a ticket to engineering (tag = "esc-to-eng" added in Intercom), assemble the customer context before the engineer picks it up:
    
    - Account: size, tier, deal stage if any, contract value
    - Recent activity: open tickets, product-usage signals from the last 14 days
    - History: prior escalations, their outcomes, who handled them
    
    Append the context as a comment on the escalated ticket. Ping the engineering on-call in #escalations with a one-line summary.
  3. 3
    Click Run now to verify

    Eyeball the first run for accuracy. Once it looks right, the routine fires on its trigger from then on. Assumes the plugin from the marquee above is already installed and connectors authorized.