Copy-paste prompt
Every Monday at 9am, read last week's support emails, group them by the underlying question, and rank the groups by how often they came up. For the top 5 groups that our FAQ at [link] doesn't cover yet, draft an FAQ entry and a saved reply based on our best past answer. Remove all customer details. Put it in a doc and link it here.
Fill in everything in [square brackets]. Put your approval rules (see the free operator's charter) above it.
What you need
- Email connector or support export
- Current FAQ page or doc
How the agent does it
- Collect the last 7 days of inbound support messages.
- Cluster them by underlying question, not wording.
- Rank clusters by count and by time spent replying (thread length).
- For the top 5, check whether the FAQ already answers it; if not, draft an entry from your past best reply.
- Draft one saved-reply macro per cluster.
What you get
A ranked table of question clusters with counts, plus draft FAQ entries and macros in a doc.
| Schedule | Mondays at 9am. |
|---|---|
| Needs your approval | Publishing FAQ changes to the live site needs approval. |
| How to check it worked | Each FAQ draft quotes the past reply it was built from; cluster counts add up to total messages. |
Common mistakes
- Leaking customer details into public FAQ text. Strip names, emails and order numbers.
Setting it up in each agent
In Grok Bot
Ask the Bot to run the prompt once and review the result. Then say "Save the process we just used as a skill called 'Weekly FAQ mining from support email'" and "Create a routine that runs the Weekly FAQ mining from support email skill mondays at 9am". Grok Bot's docs say a Bot can own up to 50 routines, schedules must be at least five minutes apart, and you should use Test run before you enable one.
In Meta Muse
Paste the prompt into Muse and ask it to remember the job and run it on the schedule. Muse asks for approval before outbound actions and purchases, so keep the 'don't send' line in the prompt. Before you rely on it, check that Muse can reach the apps this recipe needs.
In ChatGPT Dots
Give your dot the prompt as an ongoing goal and set the schedule. Dots can work between chats and ask for approval before sensitive actions. OpenAI says background work counts toward your plan's allowance, so match the frequency to what you actually need.
Run it by hand once with harmless inputs before you schedule it. A test run does real work.