Claude Code briefs twenty consulting engagements
Zille reports daily client briefs covering more than twenty active engagements, assembled from Basecamp decisions and Help Scout conversations with pointers to original records.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Read the active-thread index and relevant client records. Prepare a daily brief of outstanding tickets, unanswered messages and overnight changes. Keep internal notes separate from shareable drafts and wait for my approval before sending.
Reach people exploring AI agents
See placements →What happened
Zille reports daily client briefs covering more than twenty active engagements, assembled from Basecamp decisions and Help Scout conversations with pointers to original records.
Reported by Dave ZilleHow the agent did it
- Synchronise Basecamp posts and Help Scout conversations into separate source repositories.
- Create client profiles and summaries that point back to original records.
- Read actionable-thread indexes before opening individual conversations.
Where a person stepped in
Zille reviews edits, sends and publication. Internal and shareable views differ, and sensitive contracts and credentials remain outside the context folders.
Dave Zille starts by asking Claude Code to draft communications for one consulting client. The working system grows into daily briefs for more than twenty engagements, using local Markdown rather than a new hosted CRM.
Two repositories retain the original material: one for Help Scout conversations and one for Basecamp posts. A third organises each client’s profile, current context, recent activity and billing history. Global indexes identify actionable threads and map source IDs to files, so the agent can open the underlying conversation when a short summary is insufficient.
Scheduled runs leave a fresh brief of outstanding tickets, replies waiting on Zille and overnight changes. The same skill produces a focused client update on demand. Source repositories synchronise independently; aggregation files reference them rather than duplicating every thread. This makes it possible to investigate why a ticket remains open without discarding its history.
Zille keeps internal and shareable views separate and reviews consequential edits, sends and publication. Credentials and sensitive contracts stay outside both views. The account demonstrates an operating consultancy workflow, without an independently measured time saving. Its flat-file approach also depends on one operator’s curation; the author explicitly says it would not provide the roles and shared audit controls needed for thousands of accounts.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Blog · Dave Zille
Frequently asked questions
Why retain separate source repositories?
Each feed synchronises independently, while per-client summaries retain pointers to the full original conversations.
What appears in the daily brief?
Outstanding tickets by client and project, urgent issues, threads waiting on the consultant and overnight changes.
Does the agent send replies independently?
No. Zille says consequential edits, sends and publication require his review and approval.
More details
Remaining steps
- Generate a daily brief of outstanding tickets and unanswered threads.
- Review client-facing drafts and approve consequential actions yourself.
About this account
Claude Code follows local indexes into source records, combines project and customer context, and writes scheduled or on-demand client briefs and communication drafts.
Firsthand account by Dave Zille; original publication date verified against Primary byline Apr 30, 2026. Outcomes are the author’s reports, not independently reproduced. Suggested prompt written by Agents In Use; not the author’s original prompt.
Credit: Dave Zille
Source published 30 Apr 2026
Source checked 14 Oct 2026
Tools and topics
Claude Code · Basecamp · Help Scout · Git · Local Markdown
Consulting client briefs · Source-thread pointers · Internal client context
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.