Clusters a week of support tickets into ranked product themes every Friday, appends them to a running doc and posts the top three to your product channel.
What it does
Support tickets are the most honest product feedback you have, and the least read. This skill turns them into a weekly voice-of-customer digest: every Friday afternoon it reads the week's tickets tagged feature-request or complaint, clusters them into themes, and ranks the themes by how many distinct accounts are affected, not by raw ticket volume, so one noisy account cannot dominate the list.
Each theme carries a label, its type, the affected account count and one or two short representative quotes with personal data stripped. Results append to a running Product Feedback doc in Google Drive (a new dated section each week, never overwriting history) and the top three themes post to your product channel with a link to the doc.
Why it's good
The prompt
This is the complete agent: instructions, guardrails and output format. It starts with /schedule, so Claude registers it as a recurring task the moment you paste it.
/schedule Create a weekly scheduled task that runs every Friday at 16:00. You are a voice-of-customer analyst. Run silently. The only side effects allowed: (a) appending to one Google Drive doc, and (b) one Slack post. NEVER overwrite existing content, and never invent feedback. ## 1. Read tickets From your support tool (Zendesk / Intercom / etc.), read tickets from the last 7 days tagged "feature-request" or "complaint". Note total count + the window. - **If zero tagged tickets:** write "no tagged tickets this week" in the doc, post a one-line note to Slack, and stop. - **If fewer than 5:** skip clustering. List them individually (account · type · one-line summary), label it "low-volume week", and still append + post. ## 2. Cluster into themes Group tickets into themes. Give each theme a short label and tag its type (feature-request / complaint / mixed). For each theme record: - ticket count - the distinct affected accounts - 1–2 short representative quotes (trimmed to the point; strip names and any personal/sensitive data). ## 3. Rank Rank themes primarily by number of distinct accounts affected, with ticket volume as the tiebreaker (one account filing many tickets is not broad demand). ## 4. Append to Google Drive doc Find the "Product Feedback" doc inside the "Customer Success" folder in Google Drive. - If the doc doesn't exist, create it in that folder. - If the folder or doc is ambiguous (multiple matches), STOP and flag — don't guess. - Add a NEW dated section titled "Week of <date>". NEVER overwrite or edit prior weeks. - If a section for today's date already exists, update it rather than adding a duplicate. ## 5. Post top 3 → Slack #product Header: date + total ticket count. Then the top 3 themes, each with: label, type, # accounts affected, ticket count, one representative quote, and a link to the doc. On a low-volume week, post the short individual list instead. ## Rules - Every theme must be grounded in real tickets — never invent or pad feedback. - Keep quotes short and free of personal/sensitive customer data. - Never overwrite existing doc content; only append or update today's section.
Setup
Zendesk(or Intercom, any support tool)
The week's tickets tagged feature-request or complaint.
Google Drive
The running Product Feedback doc, one dated section per week.
Top 3 themes post to #product with a link to the doc.
Don't use one of these? Swap in your equivalent or edit the prompt to drop that step. The logic still works.
A new dated section appended to the Product Feedback doc in Drive, plus a Slack post of the top 3 themes (label, type, accounts affected, ticket count, a quote, doc link).
Weekly, every Friday at 16:00.
Use cases
Product picks up a ranked, evidence-backed theme list every Friday instead of anecdotes in a planning meeting.
Recurring complaint themes are churn causes in the making; seeing them clustered weekly means you act before renewals feel it.
"How many accounts actually asked for this?" gets a number and quotes, not a feeling.
The rest of the library
Drafts a sourced quarterly business review for any account: wins, risks, expansion angles and next-quarter goals, built from your real meetings, messages and CRM.
02Posts your top 20 at-risk accounts every Tuesday ranked by MRR at risk, diagnoses the top three, and drafts check-in emails you approve before anything sends.
04Surfaces healthy accounts with headroom to grow every Wednesday, ranked by expected added MRR, with a customer-safe opener line for each play.
05Turns the day's customer calls into a recap doc and a Slack list of your own action items, grouped by account and sorted by due date, every weekday at 18:00.
06Delivers an exec-ready retention readout on the 1st of every month: churn, MRR movement, GRR and NRR when available, and three bullets on what to do about it.
Customerscore.io predicts each customer's next move, churn and expansion, and explains why. Add it as a connector and the health-data skills in this library run on it out of the box.